Ce-doped Ni-Al LDHs Characterizations for Adsorption of Fluoride
According to the XRD investigation depicted in Fig. 1A, the as-synthesized Ni-Al-Ce LDHs exhibited distinct diffraction patterns at specific angles. The observed peaks occur at 11.47°, 23.12°, 34.94°, 39.22°, 61.0°, and 62.32°, can be related to the (003), (006), (012), (015), (110), and (113) crystal planes, respectively. These planes are associated with the hydrotalcite-like structure of the LDHs, as per the JCPDS reference code 00-015-0087. The presence of crystal planes (012), (015), (018), and (110) in the Ni-Al-Ce LDHs indicates a hexagonal coordination with rhombohedral 3R symmetry. (George & Saravanakumar 2018; Hanif et al. 2019; Jitianu et al. 2013; Wagassa et al., 2023).
In Fig. 1A, it is evident that as the amount of Ce4+ dopant increased, the diffraction peaks slightly shifted to lower 2θ, the d-spacing increased (Table S1) and the intensity of the peaks decreased. Again, upon increasing Ce4+, secondary phases appeared in addition to the LDH phases. All these may be due to that the ionic radius of Ce4+ (0.87 Å) is larger than that of Al3+ (0.39 Å) (Liao 2006).
The Ni-Al-Ce LDHs chemical compositions were also investigated using ATR-IR. The ATR-IR spectrum is shown in Fig. 1B. The intensified ATR-IR peak at 3400 cm− 1 can be attributed to the vibrational stretching of O-H groups of the layer hydroxides, and H2O molecules in interlayer (dos Santos et al. 2013; Karami et al. 2019). The peak at about 1619 cm− 1 is because of the bending vibration of the H2O molecules (Jabeen et al. 2017; Karami et al. 2019). The peak at 1361 cm− 1 may be related to vibration of the carbonate interlayer anion (George and Saravanakumar 2018; Jabeen et al., 2017; Karami et al., 2019; Lei et al. 2017). The band at about 1106 cm− 1 can be specifically assigned to Ce-O-Ce vibration in the Ce-doped Ni-Al LDHs, which is not present in the pristine LDH (Aponte et al. 2020). The intensified peak in the low-frequency region found at 556 cm− 1 can be associated with metal–oxygen-metal and oxygen-metal–oxygen bending vibrational modes as in the case of Ni-O or Al-O or Ce-O or Ni-Al-O or Ni-Ce-O vibrational modes (George and Saravanakumar 2018; Jabeen et al. 2017; Ouyang et al. 2021).
XPS was utilized to analyze the composition and electronic state of Ce4+ doped Ni-Al LDH. The survey scan XPS spectrum clearly displayed characteristic features of elements in Ni-Al LDH, and an additional signal in the binding energy range of 870 to 930 eV corresponding to the Ce 3d state. This observation indicates successful incorporation of Ce into the Ni-Al LDH matrix (Fig. 2) (Gao et al. 2021). The Ni 2p spin-orbit doublets were evident in the spectrum, with peak positions at 856.24 and 874 eV, and their shakeup signals appeared at 861.91 and 881.41 eV, confirming the presence of Ni2+ ions (Fig. 2B) (Li et al. 2018; S. Wu et al. 2017). The peaks of Al 2p were observed at 68.19 and 73.65 eV, which correspond to Al 2p3/2 and 2p1/2, respectively (Fig. 2C) (Lei et al. 2017). Furthermore, the high-resolution spectrum spanning from 870 to 925 eV in the Ce 3d region provided further evidence for the presence of Ce4+ ions in the sample (Fig. 2D) (Gao et al. 2021).
The scanning electron microscopy (SEM) images of Ce-doped Ni-Al LDHs synthesized through co-precipitation at pH 10 are presented in Fig. S1. The images reveal that the synthesized LDHs exhibit a distinct morphology characterized by irregular particle structures. However, upon closer examination, it becomes evident that these irregular particles are agglomerations of very small sheet-like grains, arranged in a layered morphology. The observed morphologies of the synthesized Ce-doped Ni-Al LDHs are consistent with those previously reported for LDHs synthesized under similar conditions (Xu et al. 2018). The irregular particle shapes and layered structure are indicative of the crystalline nature of the LDHs, with the layers stacked on top of one another. The sheet-like grains within the LDH structure likely arise from the growth of individual layers during the co-precipitation process.
Figure 3 displays TEM and HRTEM images of the Ce-doped Ni-Al LDH samples. The TEM analysis of Ce4+-containing Ni/Al LDH revealed the presence of nanoplates with a distinct morphology, forming aggregates of superimposed multilayer structure (Fig. 3A). Furthermore, the HRTEM analysis provided insight into the internal structure of the discrete nano-plates, revealing the presence of a number of nanosheets within each plate (Fig. 3B). The Ce-doped LDH exhibited a hexagonal plate-like morphology, comprised of numerous superimposed nanosheets with an average size of approximately 11.51 nm. Moreover, HRTEM examination displayed lattice fringes with d-spacing values of 0.26 and 0.2298, related to the (012) and (015) crystallographic planes, respectively (Fig. 3C). These results confirm the presence of particles of sheet-like with a structural phase of hexagonal, indicative of the layered structure of Ni-Al LDH, which aligns with the findings from XRD (Fig. 1A) and SEM (Fig. S1) investigations. The corresponding SAED patterns shown in Fig. 3D provided further insight into the Ce-doped Ni-Al LDH nanosheets. These patterns clearly demonstrated the polycrystalline nature of the Ce-doped LDH, indicating the presence of multiple crystal domains within the nanosheets. Our findings correlate with other reports (Li et al. 2018; Wagassa et al. 2023; Wu et al. 2017).
Adsorbent properties, including specific surface area and pore sizes, play a decisive role in adsorption processes. BET textural features of Ce-doped Ni-Al LDH were determined. These results of BET are presented in Table S2. They were evaluated to understand the structural properties of these LDH nanomaterials. Accordingly, the Ni-Al-Ce-0 sample exhibited surface area of 2.30 m2/g, pore volume of 0.0015 cm3/g, as well as an average pore size of 13.44 nm. On the other hand, the Ni-Al-Ce-5 sample displayed significantly higher values for surface area (25.85 m2/g) and pore volume (0.054 cm3/g) with a reduced average pore size of 4.21 nm. The observed differences in textural characteristics between the pristine and doped samples can be attributed to the presence of Ce in the Ni/Al LDH matrix. Ce is known to possess catalytic properties and can enhance the textural properties of pristine materials, like surface area and pore volume, due to its ability to modify surface chemistry and morphology. Previous studies have demonstrated the positive impact of cerium addition on the textural characteristics of catalysts (Kim et al. 2020). The significant increase in surface area as well as pore volume observed in Ni-Al-Ce LDH sample suggests a higher accessibility of active sites, which can enhance its performance in various catalytic applications. The reduction in average pore size indicates a more refined pore structure, potentially facilitating mass transfer and improving the overall catalytic efficiency (Wang et al. 2014). Therefore, these findings highlight the potential of cerium as a promoter to improve the textural properties of materials, thereby enhancing their adsorption performance.
The thermal stability of the as-prepared Ni-Al-Ce LDH was examined using TG-DCS, and its thermogram is shown in Fig. S2. The decomposition of the LDH sample as indicated in the thermogram produces a two-step mass loss peak, and a corresponding endothermic peak is observed on the DSC curve. A total mass loss of about 31% was observed. The first mass loss of about 12.5% has happened in the range of 36 to 300°C at the endothermic peak of 175 oC. This loss of mass could be ascribed to the removal of adsorbed and interlayer H2O. The second stage mass loss of about 18.5% happened at about 395 oC endothermic peak in the range of 300 to 550 oC, and this could be because of the dehydroxylation of the hydroxide layers and the decomposition of interlayer carbonate anions. Large amount of the mass loss of the synthesized LDH happened in this second stage. Above 550°C, the hydrotalcite-like structure of the LDH nanoparticles collapsed and produced mixed metal oxides. There was no significant change in mass after 550°C, indicating that the conversion of LDH to oxide was complete (Smalenskaite et al. 2017).
Adsorbent Dose, pH, Temperature, and Agitation Speed Effects on Fluoride Removal Efficiency
To investigate the effect of pH, initial concentration, contact time, adsorbent dose, temperature and agitation speed on the adsorption efficiency of fluoride on Ce-doped Ni-Al LDHs, batch adsorption technique was used. The adsorption efficiencies of LDHs were compared in the first set of tests in which Ni-Al-Ce-5 LDH was identified as the best performing LDH and it was used for further studies.
The Ce-doped Ni-Al LDH with different doses (0.25, 0.5, 0.75, 1, 1.25 and 1.5 g/L) were studied for the adsorption efficiency of F− from aqueous solution of initial concentration of 10 mg/L fluoride at 25 oC stirring at 200 rpm for a contact time of 60 min.
The result in Fig. 4A shows that adsorbent exhibit high removal efficiency of more than 92% and which marginally improved from 92.5–94.02% with the rise in the adsorbent concentration from 0.25 to 1.25 g/L, while it decreased when the adsorbent weight was further increased. As proved in previous literature, increasing the adsorbent dosage can increase the removal of fluoride ions as number of sites available for adsorbent–adsorbate interaction increase. A further increase in the concentration of adsorbent makes restrictions to mass transfer of fluoride, and consequently minimizes the number of active adsorption sites. An increase in adsorbent dosage may also lead to adsorbent aggregation, and consequently, a decrease in available adsorbent sites (Mullick and Neogi 2019).
To investigate the pH effect on fluoride adsorption efficiency, experiments were performed at pH values between 3 and 11 with an initial concentration of 10 mg/L, adsorbent dosage of 1.25 g/L, temperature of 25°C, and interaction time of 60 min. The result of pH effect on fluoride adsorption is given in Fig. 4B. As the result indicates, the removal efficiency decreased slightly from pH 3 to pH 5, then remained constant up to pH 9 and decreased beyond that. Thus, it can be witnessed that the as-synthesized LDH has a wide range of pH stability and removal efficiency.
The influence of temperature on adsorption efficiency of fluoride (10 mg/L) was tested using 1.25 g/L adsorbent at 200 rpm in the range from 25 to 55°C, and the results are presented in Fig. 4C. The results showed that throughout this range the effect of temperature was found to be negligible, and the material showed a large range of thermal stability.
The influence of stirring rate was also conducted by changing the stirring rate from 50 to 250 rpm using an incubator shaker with 10 mg/L (initial concentration), 1.25 g/L (adsorbent concentration), 60 min (contact time), and 25 ºC (temperature). Experimental results showed the removal efficiency increased significantly as the stirring speed increased up to 200 rpm (Fig. 4D). The increase in stirring rate results in the dispersion of adsorbent particles in aqueous solution which can minimize the boundary layer for mass transfer and enhances the fluoride removal efficiency.
However, after 200 rpm, the efficiency of fluoride removal decreased with increasing stirring rate. This was happened due to the higher the stirring rate, the faster the adsorbate movement and the adsorbent kinetic energy and this leads to the temporary entrapment of F− to the surface of adsorbent. Therefore, the loosely attached adsorbates began to separate from the surface of adsorbent. Again, high agitation may create a vortex which may decrease the fluoride-adsorbent interaction. A similar study by Mondal and Roy (2016) also reported that increasing agitation rate beyond 200 rpm reduced the fluoride removal efficiency by MWCNTs.
Kinetics and Isotherm Study of the Adsorption of Fluoride
A study showing the impact of interaction time on the efficiency of F− removing using 1.25 g/L LDH adsorbent in 10 ppm F− aqueous solution at agitation of 200 rpm and 25 ºC temperature is shown in Fig. S3. The results specified that the removal efficiency of F− increased in the first 60 min, and thereafter there was no significant change.
It is obvious that the existence of more adsorption sites at the start of the adsorption process allows faster removal because of the more surface area of adsorbent available for fluoride adsorption. However, further increase in contact time leads to almost a constant fluoride removal rate as the adsorption sites are occupied (El Rouby et al. 2020).
The kinetic models such as PFO and PSO were employed to investigate the performance of fluoride removing from aqueous solution over a time range of 15 to 120 min, and the results are presented in Fig. S3B and Table S3. CAVS software was employed to nonlinearly fit the models. As evidenced from the results, fluoride adsorption is slightly better fitted by the PSO kinetic model (R2 = 0.999). Thus, the results suggested that the removal mechanism was dominated by PSO kinetics of adsorption, and implying that the process was controlled by the chemisorption process (Ho and McKay 1998).
Experimental results on the influence of initial concentration of fluoride ion on its removal were tested at 10, 20, 30, 40, 50 and 60 mg/L using 1.25 g/L adsorbent at 25 oC and 200 rpm for 60 min is shown in Fig. S3C. The results revealed that the adsorption efficiency decreased with increasing initial concentration. Similar findings were previously reported indicating a decrease in uptake when initial concentration increases due to the filling of the adsorbents active sites with increasing initial concentration (El Rouby et al. 2020).
The isotherm experimental results obtained over the range of 10 to 60 mg/L concentration are given in Table S4 and Fig. S3D. Isotherm models like Freundlich, Temkin, Langmuir and Jovanovic were employed to determine the effect of initial concentration on the adsorption process. CAVS software was used for fitting the experimental results to these models. Experimental equilibrium adsorption capacity was calculated according to Eq. (1). The maximum adsorption capacity of Ce-doped Ni/Al LDH was 238.27 mg/g according to Langmuir model and 130.73 mg/g from Jovanovic model. The best fitted model provided the maximum adsorption capacity greater than the experimentally result which indicates that the tested adsorbent can hold a larger quantity of F− than those found in solution. According to the R2 values recorded in Table S4, a marginal difference was observed between the three models Freundlich, Langmuir and Jovanovic models. Freundlich was found to fit the results somewhat better than other models (Fig. S3D). Otherwise, the adsorption can be explained in terms of the three models. Thus, multilayer (physical) as well as monolayer (chemical) adsorption phenomenon can be observed between Ce-doped Ni-Al LDH and fluoride ions. Based on the dimensionless equilibrium parameter (KL) of Langmuir model values of less than unity indicates fluoride adsorption by Ce-doped Ni/Al LDH is a favorable process (Langmuir 1918).
Co-existing Ions Influence on Removal Effectiveness
The influence of Cl−, NO3− and SO42− anions at 5, 10 and 20 mg/L concentrations on the defluoridation performance of Ce-doped Ni-Al LDH is presented in Fig. S4A. The data shows that NO3− does not significantly affect the fluoride adsorption performance of Ce-doped Ni-Al LDH when present at lower concentrations, and when its concentration matches that of fluoride. This result parallels previous findings of F− removal. This may be justified by the possible chemical reaction of the AlFx complexes, the precipitation of CeF4 and its complexes being involved in the removal process. Therefore, F− has a stronger affinity for Ce-doped Ni-Al LDH adsorbents than NO3 (Wei et al. 2020).
However, even though Cl− and SO42− significantly affect the fluoride adsorption performance, the effect of SO42− is more pronounced. This was due to competitive adsorption on the active sites available on the surface of adsorbent during adsorption process. The sulfate ion has a higher anionic charge than chloride and nitrate ions, and its interfering on fluoride adsorption is possibly related to the Coulombic force. The SO42− in the solution can progressively reside in the active adsorption sites on the adsorbent surface, making the adsorbent unable to take up much of the fluoride ion (He et al. 2019; Tao et al. 2020). Therefore, the influence of these ions on defluoridation may be due to their relatively higher affinity toward LDH. As reported in a previous study, indeed LDH shows a preference for interlayer anions in the order CO32−>> F−/SO42− > Cl− > NO3− (Cavani et al. 1991; Hibino 2018).
Expected Mechanisms in Fluoride Adsorption-Post-Adsorption Characterizations
There are a number of potential mechanisms that could account for the adsorption process based on the ATR-IR analysis and the observed changes in the spectrum following the fluoride adsorption by Ce4+-containing Ni-Al LDH (Fig. S5B), whereas the adsorption of F− onto LDH exhibited negligible effects on the lattice structure of the LDH, according to the result from XRD in Fig. S5A. La-doped Li/Al LDH showed similar outcomes (Cai et al. 2018). A few methods, including surface complexation, ion exchange, hydrogen bonding, and ligand exchange, can be proposed based on the results of ATR-IR to aid the fluoride adsorption onto Ce4+-containing Ni-Al LDH.
The disappearance of OH− peaks suggests that surface complexation reactions may occur during the adsorption process. The fluoride ions (F−) likely form complexes with surface hydroxyl groups (OH−) present on the LDH surface, such as HF or H2O-F. These surface hydroxyl groups can act as binding sites for fluoride, leading to the formation of surface complexes. The formation of fluoride surface complexes can be described by the Lewis acid-base interaction, where the F− ions act as Lewis bases and the surface hydroxyl groups act as Lewis acids. This interaction involves the donation of electron pairs from fluoride ions to vacant sites on the LDH surface, resulting in the formation of stable surface complexes. Additionally, F can form surface complexes with Ce4+ as Ce4+ + 2F → CeF22+. This formation can be deduced from the absence of a peak at 1106 cm− 1 attributed to vibration of Ce-O-Ce in FT-IR (Fig. S5B). The formation of the CeF22+ complex reduces the amount of free F− in the solution, contributing to the defluorination process (Kong et al. 2019).
Another possible mechanism is ion exchange, where the fluoride ions replace other anions (such as hydroxide or carbonate ions) present in the LDH on the surface or in the interlayer. The absence of peaks corresponding to OH and CO32 in the spectrum could indicate the release of these ions upon the adsorption of fluoride. This suggests that the fluoride ions are exchanging with other anions present on the LDH surface. The Ce4+-doped Ni-Al LDH possesses a layered structure with interlayer anions, such as carbonate ions (CO32−) and OH ions. The fluoride ions can potentially replace these interlayer anions through an ion exchange mechanism. During the adsorption process, the fluoride ions can exchange places with the carbonate ions (CO32−) and OH in the interlayer spaces of the LDH. This ion exchange process occurs due to the difference in affinity between fluoride and other ions for the LDH surface (Liu et al. 2019).
The absence of peaks associated with OH in the spectrum of LDH can also suggest the involvement of hydrogen bonding in the adsorption mechanism. Fluoride ions may form hydrogen bonds with the hydroxyl groups on the LDH surface. This type of interaction can contribute to the adsorption of fluoride and the subsequent changes observed in the ATR-IR spectrum (Ren et al. 2021).
The presence of Ce4+ ions in the LDH structure could have a role in the adsorption mechanism. The F− may undergo ligand exchange reactions with the Ce4+ ions, forming Ce-F complexes. This interaction could be responsible for the changes observed in the ATR-IR spectrum, indicating the existence of fluoride on LDH surface after adsorption (Kong et al. 2019).
The layered structure of Ce-doped Ni-Al LDH nanoparticles provides a large surface area for fluoride adsorption. The mechanism might involve the interaction between F− and the active sites of the LDH surface (Ren et al. 2021).
LDH surface sites + F → LDHF (Surface adsorption)
The incorporation of Ce4+ ions into the LDH lattice enhances the surface reactivity and increases the binding sites for fluoride ions.
Reusability of Ce-doped Ni-Al LDH in Fluoride Adsorption
We used 0.1 M NaOH solution as a desorption agent to regenerate Ce4+ doped Ni-Al LDH after fluoride adsorption from aqueous solution. We performed a series of adsorption and desorption experiments with the doped LDH, and the obtained results are indicated in Fig. S4B. These results demonstrated that Ce4+ doped Ni-Al LDH could be reused up to four times without significant loss of adsorption capacity. This was because the Ce4+ dopant enhanced the reusability of the adsorbent and resulted in fluoride removal efficiencies of 92.96%, 84.5%, 78.71%, and 70.76% for the first four cycles. Therefore, Ce4+ doped Ni-Al LDH showed enhanced regeneration performance.
The regeneration performance of LDHs is a crucial factor in their practical applications for removing fluoride from aqueous solutions. The study by Tao et al. (2020) investigated the regeneration performance of Ce-AlOOH towards fluoride adsorption with 2% and 3% oxalic acid solutions. They found that Ce-AlOOH and Ce-AlOOHoa exhibited good regeneration performance with an average removal efficiencies of 76% and 70%, respectively after five consecutive regeneration cycles.