A Prospect for Environmental Remediation of Perchlorate via Cost-Effective Pinus Leaves and Dandelion Flower Powder-based Layer Double Hydride (LDH) Sorbents

2.1 Chemicals and reagents

The chemicals used in the analytical procedures, purchased from Sigma Aldrich, were of high purity and analytical grade. The list includes hydrate sodium perchlorate (NaClO₄·H₂O), magnesium nitrate hexahydrate (Mg (NO₃)₂·6H₂O), aluminum nitrate nonahydrate (Al (NO₃)₃·9H₂O), hydrochloric acid (HCl) and sodium hydroxide (NaOH).

2.2 Preparation of pinus leaf powder (P_iP) and the dandelion flowers powder (DFP)

We collected the pine leaves (Pinus strobus) and dandelion flowers from the university lawn into plastic bags. The obtained leaves were thoroughly washed with ultrapure water to remove dirt, debris, and other adhering impurities from their surface. The pine leaves were dried in an oven for 72 h at 70 °C to remove additional moisture. Then the dried leaves were pulverized into a fine powder using a mixer and grinder and were then passed through a molecular sieve of 100 mesh size (0.5 mm). The dandelion flowers were also washed with ultrapure water and were dried at a temperature of 70 °C for 24 h. The dried flowers were ground into a fine powder, using a manually operated mortar, and then sieved via of 100 mesh size (0.5 mm) sieves.

2.3 Mechano-chemical synthesis of LDHs

2.4 Structural characterization of prepared LDHs

The morphological characteristics of LDHs were analyzed under a scanning electron microscope (SEM) (Merlin, compact ZEISS 1550VP) operated at 25 kV. The X-ray photoelectron spectroscopy (XPS) (Escalab 250, Thermo Fisher, USA) was operated using an Escalab 250-spectrometer equipped with the Al-K-Alpha (1486.6 eV), 500 µm [15 kV150 W]. The analysis was extended to visualize the surface to provide useful information on the chemical state of the material’s surface. X-ray diffraction (Shimadzu DX-2700) was employed to determine the crystallography structure and orientation of the LDH materials. The operational conditions optimized for the analysis were 40 kV and 30 mA with radiation CuKa at a scanning rate of 0.12 s. The surface area was also calculated through nitrogen adsorption/desorption using a Brunauer– Emmett–Teller (BET) instrument (Tristar, Micromeritics, ASAP 2010N analyzer). The instrument was optimized at −196 °C, and the sample mass was 0.1 g. Furthermore, the amount of nitrogen absorbed was measured to evaluate the pore volume. The Barrett–Joyner–Halenda model was used to calculate the average pore size distribution. Fourier transform infrared spectroscopy (FTIR) (Nexus 670, Nicolet, USA) was also used to identify functional groups and vibration changes pre- and post-adsorption by the prepared LDHs.

2.5 Determining the LDH’s point of zero charges (pHpzc)

The point of zero charges (i.e., pHpzc) was determined using the pH drift method. The method was based on 0.01 g powdered LDHs and 100 mL of 0.01 M NaCl solution (with varying pH values ranging between 3 and 12). The mixture is taken in conical flasks, followed by shaking for 24 h, and the pH measured at the final stage was plotted against the initial pH values. The pHpzc is the point of the curve at which the initial pH equals the final pH (Srivastava et al., 2011).

2.6 The batch experiment

The batch experiment was performed in triplicate to ensure the concurrency of the results. A stock solution of 1 gL⁻¹ perchlorate was prepared with sodium perchlorate and was preserved at 4 °C for further dilution to prepare working solutions. The sample solutions prepared for the kinetic study were of 100 mgL⁻¹ strength, containing 100 mg adsorbent (i.e. representing ∼10 times the perchlorate concentration in the environment). The solutions were poured into a horizontal shaker operated at 200 rpm at a constant temperature of 30 °C. The run times selected for the kinetic study ranged between 5 min and 24 h. The resulting supernatant solution was centrifuged at 15,000 rpm for 5 min and filtered through micro syringe filtration of 0.45 μm. The isotherm study was conducted using concentrations ranging between 2 mgL⁻¹ and 500 mgL⁻¹. The Isotherm was conducted with a range of concentrations, to test the maximum adsorption capacity of the given material at given temperature time and adsorbent amount. Additionally, the process of adsorption was also tested with varying parameters, including different ranges of pH (between 3 and 12), temperature (between 20 and 50 °C), thermodynamic study, and the effect of coexisting ions.

2.7 Analysis of perchlorate-ion chromatography

The perchlorate contents in the samples were analyzed using ion chromatography (Dionex ICS 2100) equipped with an AS-20 and AG-20 guard column. The instrument was equipped with a suppressed conductivity detector and was optimized at 30 °C, 87 mA, and 35 Mm potassium hydroxide was used as an eluent. The injection volume of the samples was 200 µL.

2.8 Kinetic study and isotherm study

The kinetic models were further employed to evaluate the interaction mechanism, kinetic parameters and sorption kinetics of perchlorate on LDHs. Linear models and equations were employed:

Pseudo-first-order kinetic model:

Pseudo-second-order kinetic model:

On the other hand, the intra-particle diffusion model is described below: $${\mathrm{q}}_{\mathrm{t}}=\mathrm{k}\mathrm{i}t0.5+{\mathrm{C}}_{\mathrm{i}}$$ where ki (mgg⁻¹ min 0.5) is the intra-particle diffusion rate constant for stage i and C_i is the intercept of stage i describing the boundary layer effect in the external diffusion.

The amount of perchlorate adsorbed, q_e (mgg⁻¹), by the LDH adsorbents at equilibrium state was calculated using the following equation:

Furthermore, two isotherm models (Langmuir and Freundlich) were also considered to estimate the trends in the equilibrium adsorption data and fit the experimental data of adsorption. The models are as follows:

2.9 Thermodynamics

The thermodynamics of an adsorption process is also worth considering for assessing whether the process is spontaneous or not. Gibb’s free energy change (ΔG°) is the fundamental criterion of spontaneity. Reactions occur spontaneously at a given temperature if the ΔG° bears a negative value. The thermodynamic parameters of Gibb’s free energy change include the enthalpy change (ΔH°) and entropy change (ΔS°) for the adsorption processes were calculated using the following equations: $$\mathrm{\Delta }\mathrm{G}=-\mathrm{R}\mathrm{T}\mathrm{I}\mathrm{n}{\mathrm{K}}_{\mathrm{L}}$$ $$\mathrm{\Delta }\mathrm{G}=\mathrm{\Delta }{\mathrm{H}}^{°}-\mathrm{T}\mathrm{\Delta }{\mathrm{S}}^{°}$$ where R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), and T is the absolute temperature taken in Kelvin (K). $${\mathrm{K}}_{\mathrm{L}}={\mathrm{q}}_{\mathrm{m}\mathrm{a}\mathrm{x}}\times \mathrm{B}$$

K_L was obtained from the Langmuir isotherm, while the enthalpy change, ΔH°, and the entropy change, ΔS°, were obtained from the intercept and slope of plot ln K_L for the adsorption processes, versus temperature, 1/T (Fig. S8). The negative values of ΔG° confirm the feasibility of the process and the spontaneous nature of adsorption. The positive value of ΔH° indicates that the adsorption reaction is endothermic. The positive change in entropy ΔS° suggests likely structural changes that occurred on the adsorbent, and the randomness at the solid-liquid interface in the adsorption system increased during the adsorption process (Table 2).

3.1 LDH’s structure characterization using BET-N₂, XRD, XPS, and FTIR-spectra

To our knowledge, this is the first study to report the efficacy of perchlorate sorption using PiP- and DFP-based LDHs. The LDHs were of H-V-type isotherm at slightly high pressure (between 0.5 and 1.0 bar) Fig. S1, while the BET analysis suggested the mesoporous qualities of the LDHs. The IUPAC classification was suggestive of LDH’s mesoporous structure with a pore size between 2 and 50 nm (Rouquerol et al., 1994; see Supplementary Table S1). In this study, the BET-N₂ surface areas of the PiP, DFP, and N-LDH were 4.5 m² g⁻¹, 4.8 m² g⁻¹, and 16.7 m² g⁻¹, respectively. As PiP-LDH had the smallest surface area, indicating a considerable pore density distribution on PiP-LDH’s surface, providing more active sites for adsorption compared to other LDHs. The current results were better than a previous study (Yu et al., 2009), reporting the smallest surface of current adsorbent than alumina, silica, and zeolite, anion exchange resin, and powder activated carbon. Since the smaller surface area and pore size did not cause pore blockage, and the steric hindrance of perchlorate was less evident in small size adsorbent (Deng et al., 2010).

The interlayer structure of the LDHs was also analyzed using X-ray powder diffraction (XRD). We observed that a sharp basal reflection was apparent at a lower 2ɵ angle and that a non-basal peak was seen at a higher 2ɵ angle in the diffractograms. The consequent basal spacing was calculated from the sharp peak (0 0 3), which was 8.1 Å in the case of N-LDH. For DFP-LDH, the peaks of (0 0 6) and (0 0 9) disappeared, which was similarly observed for N-LDH (Fig. S2). With the addition of DFP, the N-LDH lost its original structure. The basal spacing for DFP-LDH was 6.1 Å, as calculated from the (0 0 3) peak. The distance between the interlayer nitrate ions was 4.1 nm in N-LDH and 3.7 nm in DFP-LDH. The PiP-LDH diffraction peaks did not follow the same pattern as N-LDH. Consequently, the addition of PiP resulted in a mixed structure of aluminum, magnesium, oxygen, hydroxide, and nitrate (Fig. S2).

XPS determined the chemical state and composition of the three LDHs. XPS analysis was performed to study the elemental chemicals’ valence states, which provide a measurement of the binding energies (BE) of the core electrons. These undergo a chemical shift after adsorption (Hayez, 2004). Initially, the analysis revealed that the LDHs were comprised of the elements magnesium, aluminum, carbon, oxygen, and nitrogen. In the pre-adsorption phase, a high-resolution spectrum of Cl_2p showed no peak of perchlorate ions (Fig. S3).

The FTIR patterns of all three LDHs before adsorption are shown in Fig. S4. The absorption spectra of three LDHs show the existence of peaks between 3400 cm⁻¹ and 3500 cm⁻¹, indicative of the hydrogen bond’s stretching vibrations of the hydroxide group in the LDH’s layer (Ferreira et al., 2006; Wang et al., 2013). A band’s presence near 1600 cm⁻¹ in all LDH’s spectra was of low intensity, suggestive of the bending vibration of the water molecules (Wu et al., 2010; Guo et al., 2012). Other bands observed near 13000 cm⁻¹ were attributed to nitrate ions in all three LDH’s interlayer. Lattice vibrational bands were evident between 600 cm⁻¹ and 400 cm⁻¹. These were ascribed to magnesium and aluminum ions. The band on the PiP-LDH spectrum located at the wavelength of 1700 cm⁻¹ arising from the C⚌O anhydride was related to the PiP (pinus powder). On the DFP-LDH spectra, the peaks appearing at 1488.1 cm⁻¹, 1270.6 cm⁻¹, and 1048.7 cm⁻¹, representing methylene, C—O—H, and C—OH stretches, respectively, were related to the contents of DFP. On the N-LDH spectra, the peak near 1700 cm⁻¹ could be caused by OH stretching.

The observed pHpzc values of the PiP, DFP LDH, were close to 6.2, and the value of N-LDH was 7.1. The pHpzc values provide an understanding of the LDH's surface behavior at different initial-pH values, which is closely related to the sorption of negatively charged perchlorate-anions.

3.2 Adsorption kinetics and isotherm study

We employed two kinetic models describing the experimental data of perchlorate adsorption. The kinetic study was carried out with an initial concentration of 100 mgL⁻¹ and 0.1 g of adsorbent. We used this concentration and amount of adsorbent in the kinetic studies for all three LDHs and observed increased adsorption in the initial stages. The reaction reached equilibrium in 20 min, 40 min and two hours for PiP-LDH, DFP-LDH and N-LDH, respectively (Table 1, Fig. S5).

A previous study established that it took six h to reach equilibrium using 1 g C-LDH, and the rate of removal was up to 95% (Wu et al., 2010). Similarly, another study reported a perchlorate removal rate of 85% using organosilica and organoclays over an equilibrium time of 2 h (Seliem et al., 2013). Another group reported 90% perchlorate removal with the use of 0.82 g montmorillonite modified with hexadecyl pyridinium chloride (Makita et al., 2014). Keeping in view the fact that the conditions in the above-discussed studies were the same, the removal rates in our study were 97% with PiP-LDH and 93% with DFP-LDHs, using 0.1 g adsorbent. These rates are comparatively higher than those reported in the previous studies. The adsorbent modified in this study (PiP-LDH and DFP-LDH) did not require acidic, thermal, or any other activation treatment before use, showing the cost- and time-effective materials. The study demonstrates the potential of natural materials such as PiP and DFP to enhance the adsorption of undesired pollutants from contaminated aqueous media within the shortest possible time.

In this study, the correlation coefficient of pseudo-second-order was higher than pseudo-first-order for all three LDHs (Fig. S6). Likewise, the calculated and experimental values of adsorption at the equilibrium state, also concurred with pseudo-second-order as summarized in Table 1, Fig. 1, indicating the suitability of pseudo-second-order in fitting the model. The models suggest that physicochemical interactions are behind the sorption of perchlorate in the case of all the LDHs prepared in this study. The notion is supported by rate constants (k₂), which were proportional to the rate of sorption. The rate constant is determined theoretically by the attributes of the given sorbate for a certain sorbent (Dai et al., 2013). The increase in the number of available sorption sites enabled the LDH’s interaction with the contaminant, via electrostatic interaction (Wang and Shih, 2011).

Close

Fig. 1 Pseudo 2nd order model of N, DFP, and PiP-LDHs.

Export to PPT

The intra-particle diffusion model (Fig. 2, Table 1) suggested that if the sorption occurred via intra-particle diffusion, then plot q_t vs. t0.5 should be linear; consequently, if the intra-particle diffusion is a rate-determining step, then the plot passes through the origin. We observed a multi-linear graph, suggestive of more than one step, controlling mechanism, for example, surface sorption and chemisorptions (Zhihao et al., 2017). It appears that sorption occurred in three steps in the cases of PiP-LDH and DFP-LDH and two for that of N-LDH. The first phase in the cases of PiP- LDH- and DFP-LDH-based sorption was surface adsorption, followed by the gradual sorption of perchlorate, in which case, the intra-particle diffusion within the pores was a rate-determining step. During the third phase of adsorption, the mechanism neared equilibrium. Similarly, the two phases of adsorption by N-LDH were surface adsorption, followed by chemisorption. The plot does not pass through the origin, suggesting that intra-particle diffusion was not a rate-determining step. Therefore, both external and intra-particle diffusion could be rate-determining steps; however, only one of them dominates at any particular time (Zhihao et al., 2017). The enormous value of the surface diffusion constant values (C_i) indicates the surface diffusion playing along compared to intra-particle diffusion as a rate-determining step during sorption (Table 1).

Close

Fig. 2 Intra-particle diffusion models of N, PiP, and DFP-LDHs.

Export to PPT

We tested the adsorption isotherms using the Langmuir and Freundlich models. The Langmuir model assumes constant adsorption energy during the reaction and that the adsorption takes place without any interactions between the adsorbate molecules. Also, the Langmuir model asserts that maximized adoption occurs when the adsorbate’s monolayer covers the surface (Shaban et al., 2018). Meanwhile, the Freundlich model supposes that the reaction is homogenous, and the increased uptake capacity results in an exponentially reduced surface binding energy. This energy forms the multilayer from the adsorbed ions (Bagherifam et al., 2014). The results of our study were in favor of the Freundlich isotherm model for describing the adsorption pattern of the LDHs, which was evidenced by the correlation coefficient values being higher than those of the Langmuir model (Table S2, Fig. 3). The calculated values of the Freundlich constant ‘n’ were 1, 1.8, and 1.5 for PiP, DFP, and N-LDHs, respectively. These are within the range of 1 to 10, suggesting a reasonable adsorption rate, which our results also showed. A value higher than 1 suggests a physical process, while the ‘1/n’ < 1 showed that the higher strength and adsorptive capacity of the adsorbent (i.e., LDH). In the case of ‘1/n’ > 1, the absorption coefficient increases with the rising concentration of the solution, leading to an increased hydrophobic surface characterization (Zhang et al., 2015; Ravi, 2014). The maximum adsorption capacity provided by the Langmuir model was 909 mgg⁻¹ for PiP, 771 mgg⁻¹ for DFP, and 246 mgg⁻¹ for N-LDH, indicating PiP-LDH’s highest adsorption capacity among three LDHs (Table S2, Fig. S7). The maximum sorption capacity of the PiP-LDH and DFP-LDH is much higher than that of powder activated carbon and anion exchange resin and usually much higher than conventional mesoporous mineral adsorbents, such as activated alumina, silica, zeolite, and montmorillonite (Yu et al., 2009). The current method provides a promising treatment of perchlorate contaminated water, especially at high contamination levels. The perchlorate distribution density on the LDH’s surface was indicative of the uptake capacity. The highest distribution density of perchlorate molecules on the LDH-surfaces can be calculated from the maximum sorption capacity and the BET surface area, based on the surface monolayer adsorption from the Langmuir model. Thus the highest numbers of perchlorate molecules per m² g⁻¹of LDH surface area (BET) were 209.30, 138.7, and 110.30 for the PiP-, DFP- and N-LDHs, respectively. Since the molecular radius of perchlorate ions is relatively large (∼2.4 Å), the exceptionally high distribution density of perchlorate molecules in the PiP and DFP-LDH were suggestive of the perchlorate anions sorption by two LDHs, by a process facilitated by interlayer anion exchange rather than surface adsorption.

Close

Fig. 3 Freundlich isotherm model of LDHs.

Export to PPT

3.3 Factors affecting adsorption

3.3.3

3.3.3 Effect of temperature

We evaluated the impact of temperature on the adsorption efficacy of the synthesized LDHs using an initial concentration of 100 mgg⁻¹. A temperature range of 20–50 °C was chosen. The adsorption rate showed little variation over the selected range. However, DFP-LDHs experienced a dramatic rise in the adsorption rate at 25 °C. It remained steady and the highest among all three LDHs across the temperature range (Fig. 4). Adsorption, in the case of N-LDH and PiP-LDH peaked at approximately 30 °C and remained steady. The PiP-LDH had a higher adsorption rate than N-LDH, although their patterns were similar. The results showed that a favorable temperature range for adsorption is between 30 °C and 50 °C. Warmer temperatures are considered to enlarge pores and increase activation of the absorbent surface. Over 30 °C, the adsorbent surface was likely to be at capacity or overcrowded by perchlorate ions, hence the steady rate of adsorption.

Close

Fig. 4 Temperature effecting perchlorate adsorption by LDHs.

Export to PPT

3.4 LDH’s dose and perchlorate uptake

In general, the use of LDHs, prepared in this study, prompted perchlorate removal that spiked from 0.025 g to 3 g, which were indicative of more active sites on the surface of LDHs (Fig. S11). In a batch adsorption system, an optimum dose of certain adsorbent is required to cut the cost; we, therefore, used an adsorbent dose of 0.1 gL⁻¹. According to the heterogeneity model, the spectrum of binding energy renders all the sites completely exposed, resulting in high surface adsorption, which is composed of active energy sites at low doses. On the contrary, the high adsorbent dose reduces the availability of higher energy sites, resulting in lowered adsorption capacity (Fleutot et al., 2012).

It was observed that increasing the dose from 0.025 g to 3 g, also increases the removal percentage of perchlorate by all LDHs, for there were more active sites on the surface of LDHs. Uptake capacities of LDHs decreases as the concentration of dose increases 0.025 g to 3 g. In a batch adsorption system, applying an optimum dose of suitable adsorbent is required to reduce the treatment cost in all subsequent tests; therefore, an adsorbent dose of 0.1 gL⁻¹ was used. According to the heterogeneity model, the surface is composed of active energy sites, since at low doses, the spectrum of binding energy renders all the sites completely exposed, resulting in high surface adsorption (Pavlovic et al., 2009). Likewise, at high adsorbent doses, the availability of higher energy sites decrease, resulting in lowered adsorption capacity (Fig. S11) Fleutot et al. (2012).

3.5 SEM image analysis of LDHs

SEM images were taken of all three LDHs, revealing the surfaces of the LDH material pre- and post-adsorption. The pre-adsorption surface of N-LDH appeared to consist of thick massive particles and an agglomerated state of particles, due to retained water molecules (Fig. 5-a). The pre-adsorption image of PiP-LDH showed a loosely bound particle structure, in which rod-like structures were embedded in the LDH, comprising of PiP (Fig. 5-b). The pre-adsorption surface of DFP-LDH was rough and compact, with the coalescence of powdery particle-like appearance. Its structural appearance was quite distinct from other LDHs (Fig. 5-c). In post-adsorption SEM images, N-LDH appeared as a tightly bound material, attributed to the adsorption of perchlorate ions. In the energy-dispersive X-ray spectroscopy (EDS) images, red-dotted particles represent the perchlorate ion adsorption. The post-adsorption image was also taken, the analysis showed that the compact and robust structure, and previously observed rod-like structures, had disappeared, which was likely a result of perchlorate embedding into the LDH structure. After the adsorption of perchlorate by DFP-LDH, the structure in the image appeared smooth and homogenous, and this is likely to be a consequence of perchlorate adsorption. The red particles in the EDS image represent chlorine and therefore show the presence of perchlorate in the structure of DFP-LDH, pinpointing the adsorption site (Fig. 5-a, b, c).

Close

Fig. 5 SEM images revealing (a) pre-adsorption and (b) post-adsorption states of LDHs.

Export to PPT

3.6 Post adsorption analysis

The mechanistic insight into perchlorate adsorption by LDHs was gained from post-adsorption characteristic XRD diffractograms (Fig. S12). XRD showed that the basal d-spacing of N-LDH changed slightly from 8.1 Å to 8.0 Å. For DFP-LDH, it shifted from 6.1 Å to 4.1 Å. The observed d-spacing changes could be explained by anion exchange and the reconstruction of the layer structure as the nitrate ions were replaced by perchlorate ions (Grover et al., 2009). Since the interlayer spacing in N-LDH decreased from 4.1 nm to 4 nm, and from 3.8 nm to 2.2 nm in DFP-LDH, the cause is likely to be ion exchange (Lv et al., 2006). Furthermore, the lessening intensity of nitrate peaks after adsorption indicated the perchlorate-LDH intercalation ion exchange is likely to have changed the charge density (Halajnia et al., 2013; Yang et al., 2014; Hu et al., 2017). PiP-LDH behaves differently after adsorption: the rebuilt PiP-LDH’s structure was similar to the rebuilt structure of N-LDH. Furthermore, the structural changes in PiP-LDH following adsorption were confirmed by FTIR spectra, revealing a perchlorate peak. The FTIR and XPS diffractograms also suggested that perchlorate replaced nitrate ions between the PiP-LDH’s layers. In FTIR, the peak related to nitrate disappeared after adsorption, suggesting anion exchange.

Post-adsorption XPS analysis (Fig. S13) revealed a high-resolution spectrum of PiP-LDH, deconvoluted into peaks with BEs at 209.3 and 209.1 eV, which confirmed perchlorate adsorption into the LDH interlayer. The peak visualized in DFP-LDH’s spectrum at BE of 208.9 eV corresponds with perchlorate adsorption. The peaks centered at BE of 209 eV in the N-LDH’s spectrum were also indicative of perchlorate surface adsorption on the LDH surface. That the elements magnesium, aluminum, oxygen, and carbon were involved in surface complexation with perchlorate was evident from the binding energy changes summarized in Table S3. The BEs of C_1s, N_1s, Mg_2p, AL_2p, O_1s showed they increased after adsorption, which increased with perchlorate adsorption. The elements mentioned above densities decreased, and BE increased due to the higher electronegativity of the perchlorate ions. The phenomenon is indicative of LDH’s role in perchlorate surface adsorption (Hudcov et al., 2018). This was also observed in the FTIR spectral analysis, where perchlorate was adsorbed on the surface through hydrogen bonding and anion exchange.

FTIR analyses showed that adsorption changed the peak position on the LDHs’ FTIR spectra (Fig. 6). The peaks located at 3400 cm⁻¹ and 1600 cm⁻¹ shifted for all three LDHs following the perchlorate adsorption. The peaks shift to a higher wavelength, indicating weak hydrogen bonding between perchlorate and hydrogen ions. Conversely, those shifting towards a lower wavelength suggested strong hydrogen bonding between pollutants and the LDH. Currently, the PiP-LDH spectra revealed perchlorate adsorption, while the peaks at 788.8 cm⁻¹ and 554 cm⁻¹ were ascribed to M-ClO₄⁻–M interactions or lattice vibrations. The peaks at 1788.6 and 1559.6 disappear after adsorption, indicating perchlorate adsorption in PiP (Wu et al., 2013). In DFP-LDH, the stronger perchlorate adsorption peak appeared at 1101.5 cm^−1, and 1781.9 cm⁻¹ indicated bonding between ClO₄⁻ and LDH. For N-LDH, the perchlorate adsorption peaks were apparent at 781 cm⁻¹ and 554.5 cm⁻¹ because of the M–pollutant–M interaction. In N-LDH, individual bonds shifted to higher frequencies, suggestive of weak hydrogen bonding occurring in this layer. The shift in OH⁻ peaks was likely nudged by electrostatic interaction, for example, hydrogen bonding between LDHs and perchlorate. The peak near 1300 cm⁻¹ was related to nitrate ions disappearing in all three LDHs after perchlorate adsorption, indicating anion exchange.

Close

Fig. 6 FTIR-spectra of LDHs revealing a shift in peak position post-adsorption stage.

Export to PPT

3.7 Mechanistic insight

In general, the ion exchange and surface adsorption facilitated the perchlorate adsorption by DFP-LDH, while that by N-LDH was apparently facilitated by electrostatic interaction. These interactions were confirmed by the results of the FTIR, XPS, and XRD analyses. Adding DFP to N-LDH induced certain structural modifications as evinced by the XRD spectra, for example, this disappearance of diffraction peaks (0 0 9) and (0 0 6) (Fig. S2). The FTIR revealed that after the addition of DFP, there was an increase in the active sites for adsorption. The peaks related to DFP-LDH, appearing at 1488.1 cm⁻¹ and 1270.6 cm⁻¹, shrunk after adsorption at wavelengths of 1408.4 cm⁻¹ and 1269.4 cm⁻¹, respectively, indicating adsorption of perchlorate on DFP via strong hydrogen bonding. Another peak located at 1048.7 cm⁻¹ disappeared after adsorption, indicating perchlorate adsorption around C—OH (Fig. S4). Furthermore, anion exchange occurred between nitrate ions and perchlorate ions, as seen in XRD analysis, where interlayer distance decreases as a result of perchlorate adsorption. The adsorption was also confirmed from FTIR spectra, revealing the disappearance of the nitrate peaks (Fig. S2).

In the case of PiP-LDH-based adsorption, a ‘rebuilding’ mechanism was involved, similar to the memory effect of C-LDH, in which CO₃²⁻ used as the interlayer in LDH (Gao et al., 2018). It disappeared after calcinated/thermal treatment and formed a mixed structure. It regained its structure after adsorption. In our study, nitrate ions were used as an interlayer anion. After the addition of PiP, a mixed structure of Mg, Al, O, OH, PiP, and NO₃⁻ formed without requiring any thermal treatment. The XRD spectra revealed (0 0 3), (0 0 6) and (0 0 9) diffraction peaks disappearing in PiP-LDH, which previously appeared in the parent, N-LDH (Fig. S2). In the post-adsorption XRD analysis (Fig. S12), a ‘rebuilt’ structure was seen, akin to N-LDH, except that the nitrate ion was replaced by perchlorate in the interlayer. Furthermore, the FTIR spectra also revealed that nitrate peaks disappeared while perchlorate adsorption peaks appeared via anion exchange. Additionally, a higher degree of anion exchange was seen with PiP-LDH compared to other LDHs, showing more adsorption capacity of PiP-LDH (Langmuir model; Table S2). Interestingly, nitrate readily exits the structural matrix; in comparison, it probably faces great hindrance when present between the layers. The active sites available to the pollutants increased with the addition of PiP. We noticed that peak at 1788.6 cm⁻¹ disappeared, indicating perchlorate adsorption by PiP. Therefore, the addition of PiP and DFP enhances the active sites and, thus, the overall adsorption capacity of LDHs.