1 Introduction

Environmental deterioration and water shortage are serious challenges in contemporary life. In particular, large amounts of sewage containing Pb2+ and organic dyes discharged by mining and textile industries enter the water circulation system, which will pose a threat to the stability of the ecological environment and human health (Zhang et al. 2020a). The World Health Organization has stipulated the maximum allowable Pb2+ concentration of 10 μg L‒1 in drinking water (Feng et al. 2022). The Pb2+, one of the most hazardous heavy metals, can enter the food chain through contaminated soil, and then impair human health by biological amplification (Kumar et al. 2020; Liu et al. 2022). Additionally, a sizable portion of water contaminants is organic dyes, which generally never degrade spontaneously due to their chemical stability. This leads to poisoning aquatic organisms and restricting aquatic plants from performing their photosynthesis (Chen et al. 2019). Therefore, an efficient and economical technology is highly desirable to decrease the excessive concentration of Pb2+ and organic dyes in wastewater.

Currently, various technologies for the removal or adsorption of Pb2+ and organic dyes have been explored, for instance, solvent extraction, membrane filtration, precipitation, and adsorption. Compared with other technologies, adsorption is one of the most widely used and economical methods because of its simple operation (Wang et al. 2020), high efficiency (Liang et al. 2022) and strong adaptability (Zhang et al. 2020a, b; Yuan et al. 2021). The research in adsorption materials is primarily focused on synthesizing and modifying carbon-based materials (Huo et al. 2021; Chen et al. 2022; Zhou et al. 2022a, b), biomass materials (Chen et al. 2021a, b), metal–organic frameworks (MOFs) (Lu et al. 2020), gel (Mo et al. 2021), and layered double hydroxides (LDHs) (Feng et al. 2022), etc. Some gratifying results have been obtained in the study of biochar as an adsorbent material (Chen et al. 2018; Yek et al. 2020; Yuan et al. 2020), but there are some problems in its use due to its cumbersome operation. Materials that are LDHs-based have been studied extensively for usage as capacitors, catalysts and adsorbents, due to their unique layered structure, abundant functional groups, high anion exchange capacity, simple synthesis, and low cost, together with other exceptional physical and chemical features (Khorshidi et al. 2022). Nevertheless, as stated, LDHs-based adsorbents suffer from an unsatisfactory adsorption capacity of Pb2+ and organic dyes on account of their undesired aggregation and dilemma for traditional synthetic methods (Wang et al. 2022a). At the same time, it is known that many metals in water are necessary and beneficial, so the enrichment is required to make subsequent centralized treatment easier. Hence, a more novel and reliable synthesis strategy remains worth pursuing to obtain LDH material with excellent adsorption capacity to remove pollutants.

Recently, the research on the three-dimensional (3D) MOFs template-directed growth of two-dimensional (2D) LDH nanosheets has been developed. This method of synthesis could overcome the problem of the stacking of the layers of LDH synthesized by the traditional method (Wang et al. 2021a). Also, the synergistic interaction of the exposed respective active sites of MOFs and LDHs could accelerate the molecular motion to enhance the adsorption capacity of both organic pollutants and heavy metal ions. Huang et al. (2022) successfully recovered Cr6+ and Pb2+ in water by synthesizing a novel sorbent Co-Al-LDH@CS/Fe3O4 using a co-metal organic framework as the precursor of Co-Al-LDH, combined with cross-linked compounds of magnetically interspersed chitosan (CS) and cysteamines. Wang et al. (2022b) used a simple one-step solvothermal method to fabricate Mg-Ni-Co LDH hollow structure (MNC HS) for the efficient adsorption of CR in water. However, to date, there are few studies on MOF/LDH nanocomposite carbonization. George G et al. investigated the adsorption of MG and crystal violet dyes using carbon-coated Zn-Al-layered double hydroxide (C-Zn-Al LDH) (George and Saravanakumar 2018). This study used the Pluronic F-127 and glucose as external carbon precursors to form a stable carbon skeleton of MOF/LDH nanocomposite after calcination. Pluronic F-127 can disperse on the surface of MOF/LDH nanocomposite as a long-chain polymer, and glucose could disperse into porous channels (Xi et al. 2018). Tris(hydroxymethyl) aminomethane buffer was also used to maintain solution pH, thus keeping the structure of MOF/LDH nanocomposite stable (Wang et al. 2022c). Additionally, the carbonization calcination can further increase the chemical and thermal stability and porosity of the internal structure because the fragile organic ligand frame of MOFs is transformed into an inorganic non-active carbon-based structure (Yu et al. 2020). Also, LDHs could be converted into corresponding bimetallic oxides (LDOs) after baking; however, their original layered structure will gradually be restored when these LDOs are exposed to water or air, which is called the memory effect (Hermosín et al. 1993). In this respect, carbonizing MOF/LDH nanocomposite is expected to strengthen the specific affinity towards organic dyes and Pb2+ and extend their practical application.

Herein, carbonizing hollow MOF/LDH nanocomposite was synthesized from the precursor, i.e., zeolitic imidazolate framework-67 (ZIF-67), or ZIF-67/LDH@C for short. The morphologies and chemical compositions of ZIF-67/LDH@C were detailly characterized. Well-integrated carbon LDH nanosheets facilitated ion sorption and desorption, and compensated for the intrinsic defects of MOF derivatives. This study aims to investigate the adsorption characteristics of ZIF-67/LDH@C as a new adsorbent for Pb2+ and organic dyes (CR and MG) and as a potential product for large-scale application. Furthermore, the addressed adsorption systems of ZIF-67/LDH@C towards Pb2+, CR and MG were evaluated based on the scientific significance of their adsorption isotherm, kinetics, and thermodynamics.

2 Experimental section

2.1 Materials and reagents

Cobalt(II) nitrate hexahydrate [Co(NO3)2 6H2O], glucose monohydrate (C6H12O6∙H2O), and sodium chloride (NaCl) were obtained from Sinopharm Chemical Reagent (Shanghai, China). Sodium tungstate (Na2WO4•2H2O, 99.5%), 2-methylimidazole (C4H6N2, 98%), and tris (hydroxymethyl) aminomethane hydrochloride (C4H12CINO3, ≥ 99%) were purchased from Aladdin Co. (Shanghai, China). Pluronic F-127 (F127, average M = 12,600), congo red (C32H22N6Na2O6S2), and malachite green oxalate [(C23H25N2)2·C2O4·2H2C2O4] were obtained from Sigma-Aldrich. Absolute ethyl alcohol (CH3CH2OH) and absolute methanol (CH3OH) were purchased from Chron Chemicals Co., Ltd (Chengdu, China). Hydrochloric acid (HCl, 37.5%) and sodium hydroxide (NaOH) were obtained from Xilong Scientific Co., Ltd (Shantou, China). The standard PbCl2 (1000 mg∙L–1) and other metal salt reagents (NiCl2, CuCl2, ZnCl2, CaCl2 and MgCl2) in analytical grade were purchased. The stock standard solutions were properly diluted to produce the working standard solutions (1000 mg∙L–1). Ultrapure water (18.2 MΩ cm) was used for solution preparation in all experiments.

2.2 Synthesis of ZIF-67 nanocrystals

The ZIF-67 nanocrystals were prepared according to the previous studies (Zhao et al. 2019). The Co(NO3)2•6H2O (3.3 g) was dissolved in 45.0 mL of absolute methanol, and a 90.0 mL of absolute methanol solution containing the 2-methylimidazole (3.7 g) was added quickly. Then, the solution mixture was aged for 24 h after stirring for 2 h at room temperature. The purple ZIF-67 powders were repeatedly washed five times with absolute methanol and separated by centrifugation. Finally, the ZIF-67 powders were washed twice with absolute ethyl alcohol and dried under vacuum for 12 h at 65 °C.

2.3 Synthesis of hollow ZIF-67/LDH nanocomposite

ZIF-67/LDH nanocomposites were synthesized by a modified method from Jiang et al. (2013). First, 90 mg of ZIF-67 precursor and different amounts of Na2WO4∙2H2O (100‒400 mg) was added to beakers containing 90 mL of anhydrous ethanol. Then, 75.0 mL of deionized water were added after five minutes of ultrasonication and the reaction was kept for about 45 min at 85 °C, until the color of suspension changed from purple to dark green. The dark green precipitates were centrifuged and washed three times with absolute ethyl alcohol. After drying under vacuum at 65 °C for 12 h, the ZIF-67/LDH with different etching degrees were collected for further experiments.

2.4 Synthesis of hollow ZIF-67/LDH@C nanocomposite

The ZIF-67/LDH@C nanocomposite utilized a referable carbon wrapping process (Jiang et al. 2015). Glucose monohydrate (25.0 mg), F127 (62.0 mg), Tris (25.0 mg), and ZIF-67/LDH (50.0 mg) were dissolved in 70.0 mL deionized water under stirring for 12 h. Then the suspension was centrifuged at 9000 r∙min‒1 for 3 min. The solid precipitate was rinsed three times with absolute ethyl alcohol, then vacuum-dried overnight at 65 °C. In order to fabricate structurally stable hollow ZIF-67/LDH@C polyhedrons, these nanocomposites were then annealed at 500 °C in flowing nitrogen for 2 h (5 °C∙min‒1).

2.5 Characterization

Scanning electron microscope (SEM) images were obtained using an S-4800 (Hitachi, Japan) at an electric voltage of 10 kV. Transmission electron microscopy (TEM) images were collected on a JEOL JEM-1230 (JEOL, Tokyo, Japan) with 80 kV. High-resolution TEM (HRTEM) and selected area electron diffraction (SAED) patterns were executed on FEI Tecnai G2 F20 at 200 kV. X-ray diffraction (XRD) patterns were measured at a powder diffractometer (Bruker D8 Advanced Diffractometer System) with a Cu Kα source at a generator current of 50 mA. Nitrogen adsorption/desorption curves were determined on nitrogen adsorption apparatus (Micromeritics, ASAP2010) to estimate the specific surface area and pore volume of materials. Thermogravimetric analysis (TGA) was performed using an STA449F3 instrument (NETZSCH, Germany). The functional groups of samples were analyzed with Fourier transform infrared (FT-IR) on a FT-IR infrared spectrometer (Vetex70, Germany) in the spectral range of 400–4000 cm–1. To characterize the elemental composition X-ray photoelectron spectroscopy (XPS) analysis was conducted on an X-ray photoelectron spectrometer (ESCALAB 250Xi). The Zeta potentials were recorded by using a Nano laser particle size analyzer (ZEN3600, United Kingdom).

2.6 Adsorption experiments

The adsorption experiments were performed by adding 5.0 ± 0.2 mg of adsorbent (ZIF-67/LDH@C) into 20.0 mL Pb2+ (co = 50 mg∙L−1), MG (co = 15 mg∙L−1) or CR (co = 15 mg∙L−1) in a glass vial. The pH value of the Pb2+ was adjusted to 3.5 with a negligible volume of 2.0 mol∙L−1 HCl or a 2.0 mol∙L−1 NaOH solution. The pH of CR and MG was approximately 6.0 and no further adjustment was required. The suspensions were then stirred in a shaking bath at a specific contact temperature (T = 298 K, 160 rpm) until adsorption equilibrium was achieved (t = 120 min). The Pb2+ solution was taken out within the fixed contact time and immediately filtered with a 0.22 μm microfiltration membrane. The concentration of residual metal ions in filter liquor was detected by atomic absorption spectrometry (Function-as-a-Service [FAAS]; TAS-990, Beijing Purkinje General Instrument Co., Ltd.). The solid phase was separated from the organic dyes solution by centrifuging at 9000 rpm for 5 min and the concentration of residual MG or CR was measured by ultraviolet (UV) spectrophotometry (UV–Vis Spectrophotometer, UV-1200, Shanghai Mapada Instruments Co., Ltd.) at 619 nm and 497 nm wavelengths, respectively. All adsorption experiments were performed under the same conditions unless otherwise stated. The adsorption capacity (qe, mg∙g‒1) and removal efficiency (R, %) were evaluated by the following equations: qe = (co-cf) × V/m and R = (co-cf)/co × 100%, where co and cf corresponded to the initial and final concentration of adsorbate (mg∙L‒1), m and V represented the mass of the adsorbent (g) and the total volume of the solution (L), respectively. The average values were taken from triplicate measurements in all experiments.

2.7 Analysis of adsorption kinetics, isotherms and thermodynamics

The adsorbents (0.005 g) were added to a 20 mL solution of Pb2+ (50 mg∙L−1) and then were shaken on a rotary shaker at 298 K for 5, 15, 30, 60, 90, 120, and 180 min, respectively. For organic dyes, 0.005 g of adsorbents were poured into a flask containing 20 mL of CR or MG (~ 15 mg∙L−1 for each) at 298 K. The flask was sealed and shaken for 10, 20, 30, 60, 90, 120, and 240 min, respectively. The supernatants were collected at different times to investigate the sorption kinetics. To further expound the adsorption behaviours of ZIF-67/LDH@C, the experimental data were analyzed by three adsorption kinetic models, the pseudo-first-order, pseudo-second-order, and intraparticle diffusion models with the following equations:

$$\mathrm{log}\left({q}_{e}-{q}_{t}\right)=\mathrm{log}{q}_{e}- \frac{{k}_{1}}{2.303}t$$
(1)
$$\frac t{q_t}=\frac1{k_2q_e^2}+\frac1{q_e}t$$
(2)
$${q}_{t}={k}_{i}{t}^{1/2}+{C}_{i}$$
(3)

where k1 (min−1), k2 (g∙mg−1∙min−1), and ki (mg∙g−1∙min−1/2) were the rate constant of the pseudo-first-order, pseudo-second-order, and intraparticle diffusion equation, respectively. qt (mg∙g−1) and qe (mg∙g−1) represented the adsorption capacity at time t (min) and the equilibrium capacity, respectively.

To discuss the interaction between pollutants and adsorbents, the adsorption isotherms of Pb2+, MG and CR were measured by increasing initial concentrations at the optimized conditions. Briefly, 5 mg of adsorbents were added to a conical flask containing 20 mL of Pb2+ with a certain concentration (31.4, 47.8, 84.6, 116.9, 170.2, 217.3, and 297.0 mg∙L‒1) at pH 3.5, respectively. Under the same amount of adsorbent and solution volume as above, the concentration of the dyes was changed as follows: CR 12.2, 23.3, 37.3, 79.1, 110, 220.8, and 367.6 mg∙L‒1; MG 13.3, 25.9, 44.8, 120.1, 166.9, 380.3, 427.3, 645.85, and 851.3 mg∙L‒1, followed by mechanical shaking for 2 h, respectively.

The experimental data were necessary to study the adsorption mechanism by applying the Langmuir and Freundlich isotherm models to analyze the sorption properties. The two isotherm models were exhibited as Equations. (4, 5):

$$\frac{{C}_{e}}{{q}_{e}}={C}_{e}\frac{1}{{q}_{\mathrm{max}}}+\frac{1}{{K}_{L}{q}_{\mathrm{max}}}$$
(4)
$$\mathrm{log}{q}_{e}=\mathrm{log}{K}_{\mathrm{f}}+\frac{1}{n}\mathrm{log}{C}_{e}$$
(5)

where Ce (mg∙L−1), KL (L∙mg−1), Kf (mg g−1 (L∙mg–1)1/n), qe (mg∙g−1) and qmax (mg∙g−1) represented the equilibrium concentration of the adsorbate solution, Langmuir adsorption equilibrium constant, Freundlich isotherm constant, equilibrium and theoretical maximum adsorption capacity, respectively.

The thermodynamic parameters of adsorption, enthalpy variation, entropy change and change in Gibbs free energy, were obtained by Eqs. (6, 7):

$$ln{K}_{0}=\frac{\Delta S}{R}-\frac{\Delta H}{R}\frac{1}{T}$$
(6)
$$\Delta G=\Delta H-T\Delta S$$
(7)

where Eq. (6) was a transformation of Van’t Hoff equation. ∆H (kJ∙mol−1) and ∆S (J∙mol−1∙K−1) were enthalpy change and entropy change, respectively. R (8.3145 J∙mol−1∙K−1) and K0 were the ideal gas constant and the adsorption equilibrium constant, respectively. ∆G (change in Gibbs free energy) could be calculated by Eq. (7).

2.8 Competition adsorption

The studies on the adsorption of mixed metal ions were performed by adding 0.005 g of hollow ZIF-67/LDH@C nanocomposite to 20 mL of the mixed metal ions (pH 3.50) containing Pb2+, Ca2+, Zn2+, Mg2+, Ni2+ and Cu2+ (~ 50 mg∙L‒1 for each), followed by similar postprocessing as that described above. Meanwhile, the 50 mg∙L‒1 Pb2+, 15 mg∙L‒1 CR, and 15 mg∙L‒1 MG were mixed as a solution to carry out experiments for interfering ions adsorption of multiple species. The adsorbents dosage and the solution volume were the same as above.

2.9 Desorption and regeneration

In a typical desorption experiment, the Pb2+-adsorbed ZIF-67/LDH@C was first rinsed with 5 mL purified water, and then the Pb2+ was desorbed with 5 mL eluent containing 0.5 mol L−1 thiourea and 1 mol L−1 HCl on the shaker (160 rpm) for 120 min. The desorption solution was filtered with 0.22 μm membrane for FAAS determination of Pb2+. The used ZIF-67/LDH@C was collected by the centrifugal method for regeneration. The regeneration of CR or (MG)-adsorbed ZIF-67/LDH@C was performed by shaking the captured material for 120 min with 5 mL of saturated sodium chloride solution with a mixture of ethanol and water (1:1, v/v). The mixture was then centrifuged, and the dye-released material was separated.

2.10 Adsorption column

Commercially available solid-phase extraction columns without adsorbents have been purchased to construct the adsorption column. A new column filled with 30 mg of ZIF-67/LDH@C was obtained for column adsorption. The Pb2+ solution with a concentration of 50 mg∙L‒1 (20 mL) was passed through the column and the effluent was collected in a glass vial. The residual Pb2+ amount was measured by FAAS. The 20 mL CR or MG (~ 15 mg∙L‒1) was also passed through the adsorption column and the residual CR or MG concentration was measured by UV–Vis, respectively.

3 Results and discussion

3.1 Characterization

3.1.1 SEM, TEM, STEM and SAED micrographs

The specific route for preparing the ZIF-67/LDH@C nanocomposite was displayed in Fig. 1. First, the ZIF-67 nanoparticle was synthesized by the coprecipitation. And then, sodium tungstate and ZIF-67 were mixed in the ethanol and H2O solution at 85 °C for 45 min. The released Co2+ ions coupled with − OH and WO42− in solution, eventually producing hollow polyhedrons composed of interconnected CoW-LDH nanosheets. Finally, the F127 and glucose monohydrate were wrapped on the surface of CoW-LDH. Subsequent calcination at 500 °C for 2 h under nitrogen atmosphere formed the ZIF-67/LDH@C.

Fig. 1
figure 1

Synthesis procedure of ZIF-67/LDH@C nanocomposite

Full size image

The microstructure, morphology, and size of ZIF-67, ZIF-67/LDH, and ZIF-67/LDH@C were characterized by SEM, HRTEM, STEM, and SAED. As a precursor, ZIF-67 had a well-shaped rhombic dodecahedral morphology and regular particle distribution (Fig. 2a and b). The diameter of most crystals was approximately 500 nm, which is consistent with those reported in the literature (Gu et al. 2020). The hydrolysis of sodium tungstate made the aqueous solution slightly alkaline, which was conducive to the formation of hydroxides on the surface of ZIF-67 by coprecipitation. Moreover, the etching degree of ZIF-67 was adjusted by the change of sodium tungstate solution concentration to partly preserve the MOFs structure of ZIF-67. Comparing Fig. 2c-f, ZIF-67 etched by 300 mg of Na2WO4•2H2O had the best morphology and microstructure of hollow cage structure and LDH nanosheets. The fine nanoflakes were evenly covered on the surface of samples, relating to facilitating the rapid diffusion of molecules and ions (Fig. 2i and j). Therefore, ZIF-67/LDH prepared by 300 mg Na2WO4•2H2O was selected as the precursor for subsequent carbon coating calcination. As seen in Fig. 2g and h, the carbonization of external carbon source (F127) and internal organic ligand (2-Methylimidazole) sacrificed partly by the lamellar of ZIF-67/LDH. However, the ZIF-67/LDH@C still maintained its complete structure and reduced cracking, which indicated that the stability of the material was increased (Xi et al. 2018). Besides, the element distribution of ZIF-67/LDH was further characterized by elemental mapping, and it was observed that carbon, cobalt, oxygen, and wolfram atoms were spread evenly on the ZIF-67/LDH structure (Fig. 2k and l), which demonstrated that the effectiveness of the free WO42‒ could insert into the interlayers of ZIF-67/LDH (Wang et al. 2022a). Additionally, fog diffraction rings and no observable crystal fringe were detected in SAED (Fig. 2m and n) of ZIF-67/LDH, which confirmed that the product was amorphous (George and Saravanakumar 2018).

Fig. 2
figure 2

SEM images of ZIF-67 (a, b) and ZIF-67/LDH structure formed by etching with different amounts of Na2WO4∙2H2O (c) 100 mg, (d) 200 mg, (e) 300 mg, and (f) 400 mg. (g, h) SEM images of ZIF-67/LDH and ZIF-67/LDH@C, (i, j) TEM images of ZIF-67/LDH and ZIF-67/LDH@C, (k) STEM image of ZIF-67/LDH, (l) carbon, cobalt, oxygen and wolfram mapping images of ZIF-67/LDH, (m) SAED analysis of ZIF-67/LDH, and (n) HRTEM spectrum of ZIF-67/LDH

Full size image

3.1.2 X-ray diffraction

The crystallographic structures of pristine ZIF-67/LDH and the carbon-coating samples (ZIF-67/LDH@C) were analyzed by XRD spectrum (Fig. 3a). The diffraction peaks of ZIF-67/LDH at 11.53°, 23.40°, 33.74°, and 59.11° agreed with the (003), (006), (009), and (110) facets, which were reflections of the hydrotalcite-like LDH phase (Hu et al. 2019). Notably, a new peak at 42.92° was added after carbon-coated annealing, and no impurity peak was detected. The positions of the distinctive diffraction peaks of ZIF-67/LDH@C all coincided with the peaks of ZIF-67/LDH, while the high intensity of (009) and (110) peaks of ZIF-67/LDH@C demonstrated that its crystallinity was higher than ZIF-67/LDH and the hydrothermal stability of LDH sheets were enhanced after carbonization. These outcomes stated that the composite ZIF-67/LDH was successfully transformed to ZIF-67/LDH@C.

Fig. 3
figure 3

(a) XRD patterns of the ZIF-67/LDH and ZIF-67/LDH@C samples, (b) Nitrogen adsorption–desorption isotherms and the corresponding pore size distribution curves (inset) of ZIF-67/LDH@C, (c) FT-IR spectra of ZIF-67, ZIF-67/LDH and ZIF-67/LDH@C, and (d) TGA analysis of ZIF-67/LDH and ZIF-67/LDH@C. XPS spectra of (e) survey, (f) C 1s, (g) O 1s, (h) N 1s, (i) Co 2p, and (j) W 4f for ZIF-67/LDH@C

Full size image

3.1.3 Nitrogen isotherm

Nitrogen adsorption–desorption isotherm and pore size distribution (PSD) curve (inset) were exhibited to examine the surface area and porous nature of ZIF-67/LDH@C. As illustrated in Fig. 3b, the isotherm could be classified as type-IV as the shape of the hysteresis ring was H3 in the range of 0.8 − 1.0 p/po. This manifested high adsorption at higher relative pressure and the presence of macropores in ZIF-67/LDH@C, which was further verified from the pore-size distribution (PSD) data obtained by the Barrett-Joyner-Halenda (BJH). The specific surface area of ZIF-67/LDH@C was calculated to be 85.03 m2 g−1 on the basis of the N2 Brunauer–Emmett–Teller (BET) method. The PSD of ZIF-67/LDH@C obtained from the BJH PSD curve, indicates that it is focused mainly on the range of 60 − 100 nm. It is concluded that ZIF-67/LDH@C nanocomposite with high BET specific surface area and the macropores structure is promising for the efficient transport for ions and dyes, leading to an excellent adsorption capacity.

3.1.4 FT-IR spectroscopy

For adsorbents, the capture of specific pollutants always relies on their surface functional groups, therefore, the Fourier-transform infrared (FT-IR) spectra of the samples were investigated and illustrated in Fig. 3c. The broad absorption band of ZIF-67 at 3469 cm−1 was assigned to the stretching mode of the characteristic − OH group and it was significantly strengthened after etching due to the formation of the layered double hydroxides nanosheets (Hou and Wu 2020), while weakened after carbonation because of the partial dehydroxylation consistent with the results of the thermogravimetric analysis (TGA). The sharp absorption band at 1641 cm−1 was ascribed to the stretching vibration of the − COOH group and it was weakened after modification. Notably, the absorption peaks of ZIF-67/LDH appearing at low wavenumbers ranging within the 500–1000 cm−1 can be related to the M–O and M–O-M lattice vibration (M represents Co or W) (Hu et al. 2019). Additionally, the ones with a wavelength ranging between 600–1500 cm−1 resulted from the bending and stretching modes of the imidazole group on ZIF-67, and the peak of C − N bond was located at 1300 cm−1 (Liu et al. 2021). Moreover, the bands at 1418 cm−1 and 428 cm−1 corresponded to the ZIF-67’s C = N and Co − N stretching vibrations, respectively, demonstrating the successful synthesis of the ZIF-67 precursor (Li et al. 2021).

3.1.5 Thermogravimetric analysis

Thermogravimetric analysis (TGA) was further implemented to evaluate the thermal properties of ZIF-67/LDH and ZIF-67/LDH@C samples, as presented in Fig. 3d and Fig S1. The first stage in the region of 100–200 °C was assigned to the evaporation of water molecules on the surface of the samples. Furthermore, the interlayer water molecules of ZIF-67/LDH were eliminated at 200–250 °C (Mohiuddin et al. 2021), while the interlayer water molecules of ZIF-67/LDH@C had vaporized after calcination, resulting in no change in this range. The combustion of ZIF-67/LDH and DE hydroxylation gave rise to the next weight loss at 250–500 °C (Olya et al. 2020). Moreover, with the increase in temperature, both the ZIF-67/LDH and ZIF-67/LDH@C nanostructures showed similar weight-loss stage in the range of 800–900 ℃, attributed to the complete collapse of hollow MOF/LDH nanocomposites or the decomposition of interlayer anions (WO42‒) (Samuei et al. 2020).

3.1.6 XPS spectra

To further confirm the chemical states of individual elements, ZIF-67/LDH@C was investigated by XPS spectra. The XPS survey spectra (Fig. 3e) confirmed the presence of the major elements (W, Co, N, O, and C) in ZIF-67/LDH@C. As presented in Fig. 3f, the C 1s XPS spectra could be deconvoluted into three peaks around 284.8, 286.3, and 288.6 eV, assigned to C − C, C − O, and C = O, respectively. The O 1s spectrum (Fig. 3g) consisted of two peaks at binding energies of 529.3 eV and 530.4 eV relating to O − C and O = C, respectively. Moreover, the N 1s spectra of ZIF-67/LDH@C (Fig. 3h) was deconvoluted into peaks at 398.1, 399.2, and 400.9 eV, attributed to N = C, N − C, and N − Co of the MOF template (Chen et al. 2021a, b; Wang et al. 2021a, b). The Co 2p core level spectrum (Fig. 3i) was resolved into two peaks at binding energies of 780.6 eV and 796.5 eV with typical shake-up satellite peaks at 785.9 eV and 802.5 eV, respectively, which were assigned to Co 2p3/2 and Co 2p1/2 declaring that the chemical state of cobalt element was Co(II) (Chen et al. 2021a, b; Poudel et al. 2021). The W 4f spectrum was decomposed into three peaks at 34.7 eV and 36.9 eV representing the 4f7/2 and 4f5/2 with a shake-up satellite peak (40.1 eV) (Fig. 3j) (Thalgaspitiya et al. 2020), which further supported the wolfram element that was successfully introduced in the process of etching ZIF-67.

3.2 Kinetics studies

Referring to the experimental results of adsorption condition optimization (Fig. S2), which was discussed in the Supplementary Material, we determined the suitable conditions for batch adsorption experiments. To explore the adsorption rate and features throughout the full adsorption process, the kinetics studies of ZIF-67/LDH@C for Pb2+ and MG and CR were conducted with different contact times. As displayed in Fig. 4a-e, ZIF-67/LDH@C had a similar adsorption process for Pb2+ and MG. The capture of Pb2+ and MG significantly rose in the first 30 min and subsequently reached the adsorption equilibrium after 60 min when the sorption efficiency nearly came to 100%. Nevertheless, compared with the capture of Pb2+ and MG, ZIF-67/LDH@C possessed a slower adsorption rate and lower adsorption amount for adsorbing CR (Fig. 4f). Because of the large number of active sites on MOF/LDH nanocomposite, Pb2+ ions and dye molecules could quickly diffuse to the interlayer and surface, while the capture rates slowed down later in the reaction due to the decrease of adsorbate concentration and the occupation of the adsorption sites.

Fig. 4
figure 4

(a) Investigation on the contact time of ZIF-67/LDH@C for Pb2+. Fitting results of the adsorption kinetics by pseudo-first-order (b), pseudo-second-order (c), and intraparticle diffusion (d) models of Pb2+ on ZIF-67/LDH@C. Investigation on the adsorption time of ZIF-67/LDH@C for MG (e) and CR (f). (g) Effect of equilibrium concentration on adsorption performance of ZIF-67/LDH@C for Pb2+. Adsorption isotherms described by Langmuir model (h), and Freundlich model (i) at 298 K. Effect of MG (j) and CR (k) equilibrium concentration on adsorption performance. Adsorption isotherms of MG described by Langmuir model (l) at 298 K

Full size image

The kinetic model fitting parameters were summarized in Table 1. The correlation coefficients of the pseudo-second-order model (R2 > 0.99) were higher than those of the pseudo-first-order model. In addition, the qe,cal of Pb2+, MG and CR calculated by the pseudo-second-order model were 156.25, 59.880 and 60.241 mg∙g−1 respectively, which agreed well with the experimental data (148.88, 59.012 and 55.096 mg∙g−1). It disclosed that the predominant adsorption mechanism of Pb2+, MG and CR on the hollow ZIF-67/LDH@C is dominated by chemisorption or strong complexation (Ho and McKay 1999; Yang et al. 2021). The k2 of Pb2+ and MG were 8.623 × 10–4 and 3.836 × 10–3 g∙mg−1∙min−1, respectively, both higher than that of CR (7.472 × 10–4 g∙mg−1∙min−1), revealing that the sorption of Pb2+ and MG on ZIF-67/LDH@C were faster than that of CR, which was consistent with the previous experimental results.

Table 1 Experimental data for the adsorption kinetics by pseudo-first-order, pseudo-second-order, and intraparticle diffusion model of Pb2+, MG and CR on ZIF-67/LDH@C
Full size table

The linear-fit plots for the ZIF-67/LDH@C’s intraparticle diffusion model did not pass through the origin during the entire time range, demonstrating that adsorption behaviors were not solely governed by intraparticle diffusion. The whole adsorption process of ZIF-67/LDH@C could be divided into two typical steps: in the first stage pollutant ions or molecules transfer from the bulk aqueous phase to the exterior surface of adsorbents, due to the high initial concentration of pollutants; the second stage was attributable to interior surface intraparticle diffusion (rate-limiting step), where pollutants drilled into the interior of nanosheets and carbon skeleton pores to gradually reach the adsorption equilibrium (Zhang et al. 2021).

3.3 Isotherm studies

As displayed in Fig. 4g-l and S6, the adsorption curves show typical adsorption behaviors with the increase of initial solution concentration, and the adsorption capacity of Pb2+, MG and CR increased successively, reaching 670.36, 2029.20 and 498.72 mg∙g−1, respectively. The linear fitting correlation coefficients of the Langmuir isotherm model (R2 = 0.9990 for Pb2+, R2 = 0.9937 for MG and R2 = 0.9963 for CR) in Table 2 are significantly higher than that of the Freundlich model, indicating that the adsorption behaviors of ZIF-67/LDH@C for the three pollutants were all consistent with the Langmuir isotherm model (Lin et al. 2021; Nazir et al. 2021). The three pollutants adsorbed on the ZIF-67/LDH@C met the adsorption of a monomolecular type, revealing that the binding sites were uniformly distributed on the two-dimensional layered surface ZIF-67/LDH@C (Saghir and Xiao 2021). The calculated maximum adsorption capacity of Pb2+, MG and GR (662.25, 1729.83 and 526.32 mg∙g−1) were perfectly in agreement with the experimental qe values. Table 3 lists the comparison of adsorption capacities of different adsorbents for capturing Pb2+, CR and MG. Obviously, the results demonstrated that the adsorption performance of ZIF-67/LDH@C adsorbent was equal to or better than that of other adsorbents, which shows the application potential of removing Pb2+, MG and CR in practical wastewater treatment.

Table 2 Adsorption isotherm parameters of Pb2+ and organic dyes at 298 K
Full size table
Table 3 Comparison of the maximum adsorption capacity of ZIF-67/LDH@C with other adsorbents for capturing Pb2+, MG and CR
Full size table

3.4 Thermodynamic studies

The study on the adsorption of Pb2+, MG and CR by ZIF-67/LDH@C at different temperatures demonstrates that the sorption percentage increases to a certain extent with the increase of temperature in the range of 273 K − 298 K, and finally tends to equilibrium (Fig. 5a). According to Fig. 5b, enthalpy change (∆H, kJ∙mol−1) and entropy change (∆S, J∙mol−1∙K−1) can be obtained from the slope and intercept. R (8.3145 J∙mol−1∙K−1) and Ko are the ideal gas constant and the adsorption equilibrium constant, respectively. ∆G (change in Gibbs free energy) can be calculated by Eq. (7). The thermodynamic values calculated have been listed in Table 4.

Fig. 5
figure 5

(a) Effect of temperature on adsorption performance to Pb2+ and (b) plot of ln(Ko) vs 1/T of Pb2+ adsorption using ZIF-67/LDH@C. (c) Adsorption toward different heavy metal ions by the ZIF-67/LDH@C. (d) Reuse of the desorbed ZIF-67/LDH@C for the cyclic adsorption. The eluent of Pb2+ contains the following species: 0.5 mol∙L−1 thiourea-1.0 mol∙L−1 HCl. (e) The Congo red-contaminated water was passed through a purification column filled with ZIF-67/LDH@C

Full size image
Table 4 Thermodynamic parameters of the adsorption Pb2+ and organic dyes on ZIF-67/LDH@C
Full size table

ZIF-67/LDH@C retained excellent capture capacity for Pb2+, MG and CR at 273 K with the removal percentage of 81.3%, 83.7% and 80.2%, respectively. The negative ∆G values might be an indication of the spontaneous nature of Pb2+, MG and CR adsorption onto ZIF-67/LDH@C (Wang et al. 2022d). The adsorption is more stable at lower temperatures, which can be explained by the fact that the values of ∆G decrease as temperature increases. All the ∆H values (314.64, 27.530 and 23.996 kJ∙mol−1) presented positive values, which showed that the adsorption process of Pb2+, MG and CR over ZIF-67/LDH@C is endothermic, so the elevated temperature can facilitate the reaction (Wang et al. 2022e). The positive values of ∆S for Pb2+, MG and CR suggest a strong favorable interaction with ZIF-67/LDH@C and imply increases in randomness at the solid/solution interface during the adsorption process (Peighambardoust et al. 2020).

3.5 Competition adsorption

Investigating the selectivity of the adsorption process in multicomponent heavy metal ion system is significant for practical applications since contaminated water commonly contains various heavy metals. The different uptake (%) for metal ions in the multi-component systems could be ascribed to their different characteristic properties, including ionic radius, hydrated radius, hydration energy, electronegativity, the hardness (Behbahani et al. 2021). Notably, it can be concluded from Fig. 5c that the adsorption of other metal ions was very low in addition to Pb2+, demonstrating that ZIF-67/LDH@C has a strong affinity of the active sites toward Pb2+. Furthermore, when dyes and Pb2+ coexist, the material still maintains efficient adsorption capacity for Pb2+, indicating that the material has certain selectivity for Pb2+, while the selectivity for dyes is not significant.

3.6 Reusability and adsorption column

Recyclability and regeneration are crucial components of adsorbents in practical applications and could be evaluated by the cyclic experiments. As displayed in Fig. 5d, the average recovery of adsorbent decreased from 99.20% to 89.87% after 5 cycles of Pb2+ capture, demonstrating the durability of ZIF-67/LDH@C. On the basis of excellent experimental results, we tested the adsorption column experiment of ZIF-67/LDH@C as packing for the quick purification of water samples. As seen in Fig. 5e, the concentration of Pb2+, MG and CR in the effluent water was dramatically reduced from 55.62 mg∙L−1, 16.60 mg∙L−1, and 15.10 mg∙L−1 to an incredibly low level of 0.00 mg∙L−1, 1.10 mg∙L−1 and 0.056 mg∙L−1, respectively, after being passed through a newly assembled adsorption column. These results clearly indicated the potential of ZIF-67/LDH@C as a filter with highly effective decontamination of actual wastewater.

3.7 Adsorption mechanism

The adsorption behaviors of Pb2+, MG and CR on ZIF-67/LDH@C are complex, including multifarious coexisting interactions. The results of the kinetics studies revealed that chemical mechanisms were mostly responsible for the adsorption on the ZIF-67/LDH@C. According to the calculated thermodynamic parameters, ΔG was negative, meaning the adsorption is spontaneous in the forward direction. The experimental results in this work showed the abundant amounts of the oxygen-containing groups on the ZIF-67/LDH@C, and the existence of these groups may play a critical role in the removal of Pb2+, MG and CR.

To further explore the adsorption mechanisms, Zeta potential, XPS and FT-IR were used to analyze the characteristics of ZIF-67/LDH@C before and after adsorption. The XPS survey spectrum (Fig. 6a) displayed additional peaks of the Pb element after adsorption, indicating the combination of Pb2+ and ZIF-67/LDH@C. As shown in Fig. 6b, the spectra of Pb 4f in ZIF-67/LDH@C were assigned to two peaks at 137.6 eV and 142.8 eV after Pb2+ adsorption, which corresponded to Pb 4f7/2 and Pb 4f5/2, respectively, while the peak separation of 5.2 eV inferred the complexation between the Pb2+ and ZIF-67/LDH@C (Kazak and Tor 2020; Cheng et al. 2022). Meanwhile, this demonstrated that the adsorption of Pb2+ by ZIF-67/LDH@C did not alter the valence state of Pb. The spectra of C 1s (Fig. 6c) in ZIF-67/LDH@C were deconvoluted into three peaks at 284.8 eV, 285.4 eV and 289.1 eV ascribing to C − C, C − O and C = O, respectively (Cheng et al. 2022). After the Pb2+ adsorption, the peak of C − O shifted to lower binding energy from 286.3 eV to 285.9 eV. On the contrary, the band of C = O shifted to higher binding energy as a result of the reduced electron densities (Yan and Li 2021). Figure 6d presents the O 1s spectra of ZIF-67/LDH@C after adsorption, which manifested three peaks with a binding energy of 529.4 eV (assigned to O = C), 530.6 eV (assigned to O − C) and 531.6 eV (assigned to O-Pb). When compared to the peaks in ZIF-67/LDH@C before Pb2+ adsorption, a new peak at 531.6 eV was found, which suggested that oxygen atoms were combined with Pb atoms to form Pb–O bonds (Yan and Li 2021).

Fig. 6
figure 6

XPS spectra of (a) survey, (b) Pb 4f, (c) C 1s, and (d) O 1s for ZIF-67/LDH@C after Pb2+ adsorption. (e) Zeta potential of ZIF-67/LDH and ZIF-67/LDH@C, and (f) FT-IR spectra of ZIF-67/LDH@C before and after adsorption of MG and CR. (g) Illustration of the mechanism of Pb2+, MG and CR capture by ZIF-67/LDH@C

Full size image

Furthermore, according to Zeta potential (Fig. 6e), ZIF-67/LDH and ZIF-67/LDH@C were always negatively charged between pH 1 and 7, which manifests another reasonable mechanism for attracting positively charged Pb2+. In short, surface coordination and electrostatic interaction are the primary mechanisms for Pb2+ adsorption on ZIF-67/LDH@C. These results further support the mechanism that the removal of Pb2+ by ZIF-67/LDH@C is mainly controlled by electrostatic interaction between O-containing functional groups and Pb2+ and partial formation of coordination bonds between oxygen groups and Pb2+ (Zhou et al. 2022a, b). The negatively charged surface of ZIF-67/LDH@C also provides strong evidence for electrostatic interaction between adsorbent and cationic dye MG (Liu et al. 2019). This might be the major cause of the higher maximum adsorption capacity and faster achievement of adsorption equilibrium for MG, compared with CR. However, the adsorption selectivity of Pb2+ should differ from the above mechanism analysis. Oxygen-containing groups have the ability to absorb most heavy metal ions. However, this study has selective adsorption capacity for Pb2+, which may be attributed to the principle of hard and soft acids and bases (HSAB) (Li et al. 2015). The strength of covalent bonds between Pb2+ and oxygen/nitrogen atoms in the complexes significantly differed among heavy metal species based on the HSAB principle, essentially determining the affinity and adsorption selectivity of Pb2+ on ZIF-67/LDH@C.

Figure 6f displays the FT-IR spectra of the ZIF-67/LDH@C for MG and CR both before and after adsorption. The structures of cationic MG and anionic CR are displayed in Table S1. According to Fig. 6f, after the ZIF-67/LDH@C adsorption of MG and CR, the peak around 3472 cm−1, which was associated with the stretching vibration of − OH functional groups, changed to 3459 cm−1 and 3452 cm−1, respectively, in comparison to the original ZIF-67/LDH@C (Xiao et al. 2020). Moreover, the peak kurtosis of hydroxyl decreased, and the peak deformation was wide and flat after adsorption, implying that presumably adsorption mechanisms for ZIF-67/LDH@C to dyes are Dipole–dipole H-bonding (between hydroxyl groups of adsorbents and nitrogen on dyes) and Yoshida H-bonding (between the aromatic ring of dyes and oxygen on − OH) (Wen et al. 2022). The conjugation interaction (π-π interaction) between ZIF-67/LDH@C (imidazole aromatic ring in ZIF-67) and dye molecules (conjugate aromatic structure moieties) is suggested by the fact that the adsorption peak at approximately 1585 cm−1, attributed to the stretching vibration of aromatic ring shorted to 1589 cm−1 and 1593 cm−1 after MG and CR adsorption, respectively (Arabkhani and Asfaram 2020; Yan and Li 2021). The present ZIF-67/LDH@C contains OH group from their precursors. The positive-charged cationic dyes (MG) have the electrostatic adsorption with OH group in the adsorbent, resulting in the high adsorption capacity. Additionally, MG contains a phenyl ring without dimethanamine group on it. The phenyl ring acts as an electron with drawing group when the OH group exist in the adsorption system, which will enhance the interaction between MG and the adsorbent (Lin et al. 2021). The adsorption of cationic dyes (CR) may occur on the outer planner surfaces and in the interlayers where OH group are present (Wu et al. 2019; Li et al. 2020). Therefore, the adsorption capacity of the adsorbent towards MG is higher than that of CR due to multiple synergistic effects. From the above experimental results, the different capture mechanisms of Pb2+, MG and CR on ZIF-67/LDH@C can be explained (Fig. 6g).

4 Conclusions

To remove the pollution of Pb2+ and organic dyes, a novel carbonizing hollow MOF/LDH nanocomposite named ZIF-67/LDH@C was effectively fabricated by etching, co-precipitation and carbonization calcination. Multiple characterizations techniques, the pseudo-second-order kinetic model and the Langmuir isothermal model manifested the composite multi-dimensional structure, multiple adsorption sites and adsorption behaviour, respectively. Attributed to the synergy between ZIF-67 and carbonizing LDH (electrostatic attraction, π-π interaction, hydrogen-bonding interaction and surface complexation), ZIF-67/LDH@C exhibits ultrahigh adsorption capacity, wide ionic strength applicability, and exceptional selectivity for Pb2+. However, it also has a certain ability to remove MG (qmax 1729.83 mg∙g−1 at 298 K) and CR (qmax 526.32 mg∙g−1 at 298 K) from individual species in the aqueous solutions. The satisfactory results further imply that ZIF-67/LDH@C has the potential to remove Pb2+, CR and MG from wastewater in practical applications. Thus, this work will help formulate an applicable strategy for the treatment of Pb2+ and organic dyes in wastewater based on the carbonising MOF/LDH nanocomposite adsorbent.