Nickel-iron hydrotalcites (NiFe-LDH) with various [Ni]/[Fe] ratios were prepared using co-precipitation method (NiFe-LDH CP) and hydrothermal method (NiFe-LDH HT), respectively. The crystal structure and chemical composition of NiFe-LDHs were characterized by X-ray diffraction, Fourier transform infrared spectroscopy, thermogravimetry and differential scanning calorimetry (TGA-DSC) and scanning electron microscopy, and NiFe-LDHs were occupied as catalysts to establish heterogeneous Fenton systems for the degradation of methylene blue (MB). The catalytic potential of the catalysts was investigated through cyclic voltammetry analysis. The effects of the dosage of catalyst, initial solution pH and the amount of hydrogen peroxide on the removal of MB were investigated. The results showed that the optimum ratio of [Ni]/[Fe] in the preparation of NiFe-LDHs was 3. NiFe-LDH HT is much smaller and in uniform particle size, with better redox reversible characteristic and catalytic potential. The optimum conditions for the removal of MB catalyzed by NiFe-LDH CP and NiFe-LDH HT were both determined to be 0.15 g/L catalyst, 0.88 mmol/L hydrogen peroxide at pH 2, under which the chemical oxygen demand (COD) removal were 58.96% and 67.87%, respectively. The maximum apparent generation efficiency of ·OH were 46.21% and 49.24%, and NiFe-LDH CP and NiFe-LDH HT were verified to be of high stability.
Dyes are usually of high toxicity and chemical stability, and cannot be effectively degraded by conventional biological remediation processes (Qin et al. 2015; Babu & Murthy 2017). Advanced oxidation processes (AOPs) characterized by the generation of hydroxyl radicals (·OH) are considered a potential alternative in the treatment of refractory organics, due to that ·OH is of high oxidation potential (2.8 eV) to oxidize organic pollutants quickly and non-selectively (Inchaurrondo et al. 2016). Fenton system consist of ferrous ion (Fe2+) and hydrogen peroxide (H2O2) is considered to have a fast reaction rate, easy operation and high efficiency, but with a narrow pH window. Also, Fe2+ involved may influence water quality and iron sludge generated as an additional environmental contaminant (Wang et al. 2012). In order to overcome the problems mentioned above, heterogeneous catalysts involved in Fenton-like reactions have been developed (Li et al. 2015; Chen et al. 2017).
Layered double hydroxides (LDHs) as a two-dimensional (2D) anionic clay is usually expressed by the formula [M2+1-xM3+x(OH−)2]x+(An−)x/n·mH2O, in which M2+ and M3+ represent divalent and trivalent metal cations (Iyi et al. 2008). LDHs were of large surface area and high stability for the layered structure, and those that contained iron were verified to be of good photo-catalytic capacity and have been successfully used as catalyst in photo assisted reactions (Tang & Liu 2016). Also, N. T. Thao et al. revealed that Fe-Mg-Al LDHs showed good catalytic capacity in Fenton system and in our previous studies; FeFe-LDH was verified to be an efficient catalyst in a heterogeneous Fenton process for the degradation of dye and iron was the active site for the catalytic generation of hydroxyl radicals contributing to the oxidation of methylene blue (MB) (Wang et al. 2014). However, the relationship among the morphologies of LDHs, the preparation methods and the catalytic capacity were still unknown.
Herein, NiFe-LDHs were prepared using co-precipitation and hydro-thermal methods, respectively. The LDHs as-prepared were characterized using X-ray diffraction (XRD), scanning electron microscopy (SEM), Fourier transform infrared (FT-IR), thermogravimetric analysis and differential thermal analysis (TGA/DSC). Redox reversible property of LDHs, and influences factors including initial solution pH, dosage of catalyst and H2O2, reaction temperature were investigated, and the stability of LDHs were also tested.
Material and methods
All chemical reagents were of analytical grade and were used without any further purification.
Synthesis of NiFe-LDH
Synthesis of NiFe-LDH by co-precipitation method
The NiFe-LDH with CO32− interlayered synthesized via co-precipitation method was prepared as follows: Typically, an aqueous solution (A, 100 mL) containing Ni(NO3)2·6H2O and Fe(NO3)3·6H2O was prepared (Ni/Fe = 2, 3, 4 and 6, [Ni2+] + [Fe3+] = 1.2 mol·L−1). Meanwhile, an aqueous solution (B, 100 mL) containing NaOH ([OH−]/([Ni2+]+2[Fe3+]) = 1.6) and Na2CO3([CO32−] = 2[Fe3+]) was prepared. Solutions (A) and (B) were added simultaneously into a 500 mL flask under vigorous stirring at a constant rate at room temperature. The suspension was aged at 358 K for 3 h and then the solid phase washed with deionized water to eliminate excess ions. Finally, the resulting slurry was dried at 358 K overnight.
Synthesis of NiFe-LDH by hydrothermal method
An aqueous solution (80 mL) containing Ni(NO3)2·6H2O, Fe(NO3)3·9H2O and urea ([Ni2+] + [Fe3+] = 0.1 mol/L, Ni/Fe = 3, [Urea]/([Ni2+] + [Fe3+]) = 6.6) was prepared and transferred into a Teflon reactor contained in a stainless steel autoclave at 423 K for 24 h. After been cooling down to room temperature, the solid product was washed to neutral with deionized water, dried at 373 K overnight after centrifugation and then kept sealed after being ground.
Electrochemical characterization of NiFe-LDHs
NiFe-LDH was mixed with acetylene black and adhesive (8:1:1), after being dissolved in ethanol and treated with ultrasonic for 30 min, it was film coated on nickel foam as the working electrode for the electrochemical test. Cyclic voltammetric (CV) measurements on NiFe-LDHs were performed on a Corrtest electrochemical workstation equipped with a reference electrode (SCE) and an auxiliary platinum electrode at the scan rate of 50 mV/s. 0.1 M Na2SO4 was used as electrolyte in CV measurement. Before the experiments, the dissolved oxygen in solution was purged out with N2.
All the experiments were performed in a 250 mL conical flask under constant magnetic stirring at room temperature (25 ± 5 °C). Typically, 200 mL of 10 mg/L MB solution was adjusted with 0.1 mol/L H2SO4 and 0.1 mol/L NaOH. Then, with a given amount of catalyst added, H2O2 was introduced to trigger the degradation experiments. Influences of the dosage of NiFe-LDH (0.25 g/L −1.25 g/L), the dosage of H2O2 (0–1.32 mmol/L) and initial solution pH on the degradation of MB were investigated. Solutions were sampled out at regular intervals and detected immediately after filtration.
The crystal phase and crystallinity of NiFe-LDH was characterized using an XRD diffractometer (XRD-6000, Shimadzu) using Cu Kα (40 kV, 40 mA) as an X-ray source at a scanning rate of 4°/min in the range of 10° to 80°. The surface morphology was observed by SEM with a FEI XL30F microscope operating at 20 kV. Functional groups were characterized by Fourier transform infrared spectrometry using a NEXUS (Thermo-Nicolet) spectrometer using the KBr pellet technique in the range of 4,000–500 cm−1. Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were performed on a Mettler Toledo system at air atmosphere with a heating rate of 10 °C/min in the temperature range of 30–600 °C.
The concentration of MB was monitored at 660 nm, and all the spectrophotometric measurements were performed on a UV-755B UV-Vis spectrometer (Shanghai Youke Instrument Co. Ltd). Chemical oxygen demand (COD) was determined by a closed reflux colorimetric method (Anotai et al. 2006). Samples were treated using solid phase extraction method after filtration and then eluted with CH3CN. The samples were concentrated with N2 and then dissolved in dichloromethane for GC–MS (Thermofisher (ISQ)) equipped with a fused silica capillary column (TG-SQC 15 m, I.D.: 0.25 mm, film: 0.25 m) measurement. MB degradation products were also identified using a UPLC/MS system consisting of a Waters Acquity UPLC (PDA detector) equipped with Waters BEA C18 column, 2.1 mm × 100 mm, 1.7 μm, and a Waters Quattro Premier XE Mass Spectrometer equipped with an atmospheric pressure ionization probe. The eluent consisted of methanol (A), and water with 0.1% formic acid (B) served as mobile phase in a gradient mode (10% A at 0–5 min, 10–90% A at 5–15 min, 90% A at 15–20 min) at a flow rate of 0.2 mL/min.
RESULTS AND DISCUSSION
Morphology and structural characterization
SEM images of NiFe-LDH prepared using co-precipitation method (NiFe-LDH CP) (a) and hydrothermo method (NiFe-LDH HT) (b), as shown in Figure 1, revealed that NiFe-LDH CP is composed of large-scaled sheet-like morphology, while NiFe-LDH HT perfect spheres with narrow size distributed at around 500 nm, which is much smaller. The images also clearly illustrate the differences between the surfaces of the porous NiFe-DLH CP and the smooth NiFe-LDH HT.
The phase and purity of the NiFe-LDHs prepared were characterized, and XRD patterns of NiFe-LDH CP with different ratios of nickel on iron (Figure 2(a)) and NiFe-LDH HT (Figure 2(b)) revealed that all diffraction peaks at 11.36°, 23.04°, 34.56°, 38.86°, 46.30°, 60.36°, 61.52° can be indexed to (003), (006), (009), (012), (015), (018), (100) and (113) planes of a pure NiFe-LDH (JCPDS Card, No. 33-0429). There are good multiple relations among the diffraction peak of crystal plane (003) and the d value of the two advanced diffraction peaks, indicating an excellent layer structure. No peaks of any other undesirable phases are detected, suggesting the high purity and good crystallinity of the products. The peaks at around 11.36° (2θ) are attributed to the basal space where the pillaring of metal cations took place. The basal spaces, d, derived from Bragg's law are summarized in Table 1.
As depicted in Figure 2 and Table 1, the basal space of NiFe-LDH was influenced by the ratio of [Ni]/[Fe] and preparation method. That of NiFe-LDH HT (7.7272 Å) is much wider than of NiFe-LDH CP, and the basal space depend on the ratio of [Ni] on [Fe] during the synthesis process. The basal space increased from 7.4980 Å to 7.6191 Å as the ratio of [Ni]/[Fe] increased from 2 to 3, but decreased as the ratio further increased. Also, as shown in Figure 2(a), the peaks of NiFe-LDH at the ratio of Ni/Fe = 3 are more prominent and the two peaks indexing to (100) and (113) at around 61–62 ° are more discernible. The width of the basal space is probably decided by the density of atoms arranged in crystal face (110) which was reflected by parameters a and d (110), which depend on the molar ratio of Ni to Fe. The radius of Ni2+ is larger than that of Fe3+, and Ni2+ charged less. As Ni/Fe mole ratio increased, density of atom arranged on crystal faces decreases, and the distance between metal ions in the adjacent hexagonal cells increases, so d(110) and a increased. Meanwhile, the maximum d(003) and the parameter c can be obtained with Ni/Fe = 3, which is consistent with the regularity of the structure of NiFe-LDHs with mole ratio changes.
Thermal stability of NiFe-LDH CP and NiFe-LDH HT were studied by TGA/DSC under air atmosphere and the results are shown in Figure S1(a) and S1(b) (available with the online version of this paper). The TGA curves in Figure S1(a) exhibit two slopes between 35–172 °C (wt.10.74% loss) and 172–274 °C (wt.19.49% loss). This observation is in good agreement with DSC curve that exhibit endothermic peaks at approximately 175 °C and 276 °C, which were likely related to the loss of bound water interlayered and –OH on host layer, CO32− interlayered, respectively. The layer structure damaged to form multi-element metal oxide. The same process occurred to NiFe-LDH HT (Figure S1(b)) between 35–181 °C (wt.9.03% loss) and 181–292 °C (wt.16.73% loss). And NiFe-LDH HT is of a little higher thermo-stability than NiFe-LDH CP. From the results depicted in Figure S1, we also learned that NiFe-LDH CP contained more CO32− interlayered.
Figure S2(a) and S2(b) (available online) present the FT-IR spectra recorded in transmission mode for NiFe-LDH CP and NiFe-LDH HT. As shown in Figure S2, the bands at 3,409 cm−1 (Figure S2(a)) and 3,416 cm−1 (Figure S2(b)) is due to the O-H stretching in H2O interlayered, smaller than that free O-H (3,600 cm−1), which maybe resulted from the influence of a hydrogen bond between H2O interlayered and hydroxyl group on host layer. Bands ranged in 1,500–1,635 cm−1 are vibrations in O-H in water interlayered, and 1,357 cm−1 (Figure S2(a)) and 1,362 cm−1 (Figure S2(b)) are asymmetric stretching vibration of C-O in CO32−, which are also much smaller than that CO32− (1,429 cm−1) in CaCO3 for the influence of hydrogen bond. The typical bands for layers appear between 1,000 and 400 cm−1.
Electrochemical characterization of NiFe-LDHs
To evaluate the reversible redox character of NiFe-LDH CP and NiFe-LDH HT, cyclic voltammetry (CV) measurements were carried out in 0.1 mol/L Na2SO4 aqueous solution using a three-electrode system. Figure 3(a) and 3(b) present the CV curves of NiFe-LDH CP and NiFe-LDH HT at a scanning rate of 4 mV·s−1. As depicted, ip,c/ip,a of NiFe-LDH HT is more approaching 1 than that of NiFe-LDH CP, and the potential difference (ΔEa,c) of NiFe-LDH HT was calculated to be 22.35 mV, while that of NiFe-LDH CP was 40.61 mV. Thus, NiFe-LDH HT is of a much better reversible redox character, and one electron is transferred as Fe(III) in NiFe-LDH HT reduced to Fe(II) and then regenerated (Gosser 1993; Wang et al. 2013). The appearance of reduction current peak at about 0.3786 mV was due to Na+ in solution.
Removal of MB in different systems
Combined with the observations above, possible degradation mechanism of MB with initial conditions of pH = 2, 0.88 mmol/L [H2O2], 0.50 g/L LDHs in 200 mL10 mg/L MB were investigated. KI and NaN3 were occupied as scavengers for oxygen and ·OH in solution, ·OH generated on catalyst surface, respectively, to make the active radicals involved clear, and the results are shown in Figure 4. The degradation of MB catalyzed by NiFe-LDH CP and NiFe-LDH HT were both inhibited with the addition of scavengers and the influence of KI was obviously strong. The results demonstrated that all the three radicals contributed, but ·OH generated on catalyst surface played the leading role in the degradation of MB.
Effect of parameters on the degradation of MB
Effect of initial pH
The effect of initial solution pH on the degradation of MB in Fenton-like systems catalyzed by NiFe-LDH CP and NiFe-LDH HT were investigated and the results were shown in Figure 5(a) and 5(b). It is clearly observed that initial solution pH played an important role in MB removal. The MB removal efficiencies increased as the reactions programmed, and performed excellent at pH 2 with 100% degradation efficiency obtained at 60 min. The removal efficiency dropped as initial solution pH departed from 2, and 42.71%, 3.47% and 2.99% removal efficiencies were detected at 60 min at initial pH 1, 3 and 4. At lower pH, the removal efficiency of MB dropped, due to the scavenge of ·OH by hydrogen ions. At higher pH, concentrations of dissolved iron in solution was much lower, dissolved iron precipitated as Fe(OH)2 and Fe(OH)3 in solution which are unavailable for Fenton process (Ting et al. 2009). Thus, pH 2 was the optimal condition for MB removal due to more generation of ·OH from the catalysis degradation of H2O2 by NiFe-LDH CP. The removal of MB catalyzed by NiFe-LDH HT shown in Figure 5(b) was in the same trend with that shown in Figure 5(a), but the reaction was faster and the removal efficiency of MB at the given intervals was calculated higher. The reason maybe that NiFe-LDH HT was in much smaller particle size to have larger surface area, leading to a sufficient contact with H2O2 in solution to form more ·OH for the degradation of MB (Huling et al. 2009); on the other side, NiFe-LDH HT with larger surface area may have higher adsorption capacity to accelerate the degradation of MB (Hu et al. 2011).
Effect of H2O2 concentration
Effect of catalyst dosage
Effects of catalyst dosage on the degradation of MB catalyzed by NiFe-LDH CP and NiFe-LDH HT are shown in Figure 7(a) and 7(b), respectively. It could be seen that the removal efficiency of MB enhanced with the increase of catalyst dosage from 0.05 g/L to 1.0 g/L, for more catalyst provides more active sites to accelerate the generation of ·OH (Zhang et al. 2014). Also, larger surface area and more active sites were provided for the adsorption of H2O2 and catalytic degradation of MB. However, as the dosage of NiFe-LDH further increased from 1.0 g/L to 1.25 g/L, the removal kinetic of MB decreased. An excess amount of catalyst would have a negative effect on MB removal due to the consumption of hydroxyl radicals (Elshafei et al. 2010). 0.15 g/L NiFe-LDH was chosen, and almost 100% degradation of MB can be achieved.
COD removal in MB aqueous solution
The mineralization of MB were investigated and the original COD values of MB and that after treated in Fenton-like reactions catalyzed by NiFe-LDH CP and NiFe-LDH HT under optimal condition are shown in Figure 8. As shown in Figure 8, the COD of MB before and after treated in Fenton-like reactions catalyzed by NiFe-LDH CP (A) and NiFe-LDH HT (B) were detected to be 164.0, 67.3 and 52.7 mg/L, respectively, and the COD removal efficiencies were calculated to be 58.96% and 67.87%. The results of the COD removal were consistent with the removal of MB, and NiFe-LDH HT is verified to be more efficient.
Apparent generation of ·OH
The presence of CO32− and HCO3− were reported to compete for ·OH in Fenton system under acid condition, inhibit the degradation of MB (Sun et al. 2007; Ou et al. 2013), with initial solution pH 2, and H2CO3 generated instead of CO32− and HCO3−. This is also the reason that heterogeneous Fenton catalyzed by NiFe-LDH HT performed much better than that catalyzed by NiFe-LDH CP (consistent with Figure S1 that NiFe-LDH CP contained more CO32− interlayered) and pH 2 was the optimum condition for the degradation of MB.
Fenton was reported to be of strong oxidation potential, and pollutants in Fenton systems degrade into CO2, H2O and inorganic salts, theoretically. Intermediates and products after being treated in Fenton-like processes catalyzed by NiFe-LDH CP and NiFe-LDH HT were analyzed, and benzothiazole as a special intermediate was detected in both systems (Table 2). Products with larger molecule weights, organic acids and HCOH with lower molecule weights were also detected.
NiFe-LDH CP and NiFe-LDH HT were synthesized using co-precipitation method and hydrothermal method. NiFe-LDH CP were sheet structures in large scale, and NiFe-LDH HT was of ball-like structure in much smaller scale. The basal space of NiFe-LDH CP increased with the ratio of [Ni]/[Fe] increased, and the optimal ratio was determined to be 3. NiFe-LDH HT contained less CO32− interlayered and its basal space was larger than that of NiFe-LDH CP. NiFe-LDH HT is of better reversible redox character and higher catalytic capacity for the generation of hydroxyl radicals. The optimal conditions for the degradation of MB in heterogeneous Fenton-like systems catalyzed by NiFe-LDH CP and NiFe-LDH HT were both determined to be pH 2, 0.15 g/L catalyst and 0.88 mmol/L H2O2, and almost 100% removal of MB can be reached in both heterogeneous Fenton systems. NiFe-LDH CP and NiFe-LDH HT are of high stability and with much less iron loss compared to the traditional Fenton.
This study was supported by the project of the National Natural Science Foundation of China (Grant No. 51608412), the Natural Science Basic Research Plan in Shaanxi Province of China (Grant No. 2016JQ2034).