Polyethylene glycol functionalized Fe₃O₄@MIL-101(Cr) for the efficient removal of heavy metals from Ligusticum chuanxiong Hort

2.1 Chemicals and materials

Iron chloride hexahydrate (FeCl₃·6H₂O, 99.0 %), chromium(III) nitrate nonahydrate (Cr(NO₃)₃·9H₂O, 99.0 %), dopamine hydrochloride (C₈H₁₁NO₂·HCl, 98.0 %), tris(hydroxymethyl) aminomethane (C₄H₁₁NO₃, ≥99.8 %), sodium hydroxide (NaOH, 97.0 %), terephthalic acid (C₈H₆O₄, 99.0 %), and phosphoric acid (H₃PO₄, ≥99.0 %) were purchased from Aladdin Biochemical Technology Co., ltd. (Shanghai, China). Sodium acetate (C₂H₃NaO₂, 99.0 %), ethylene glycol (C₂H₆O₂, 98.0 %), and N, N-dimethylformamide (DMF, 99.5 %) were from Macklin Biochemical Co., ltd. (Shanghai, China). Polyethylene glycol (PEG, 600 g/mol) was purchased from J&K Scientific Technology Co., ltd. (Beijing, China). Hydrochloric acid (HCl, 36.0 ∼ 38.0 %) and ethanol (C₂H₅OH, 99.7 %) were supplied by Chron Chemicals Co., ltd. (Chengdu, China). Nitric acid (HNO₃, 70 %) was purchased from Crystal Clear Chemical Co., ltd. (Suzhou, China). Acetonitrile (ACN, HPLC grade) was purchased from Sigma-Aldrich (Shanghai, China). Distilled water was purified by a Milli-Q system (Millipore Co.).

Arsenic standard stock solution (As⁵⁺, 1000 μg/mL, 2 % HNO₃), nickel standard stock solution (Ni²⁺, 1000 μg/mL, 5 % HNO₃), zinc standard stock solution (Zn²⁺, 1000 μg/mL, H₂O), cadmium standard stock solution (Cd²⁺, 1000 μg/mL, 5 % HNO₃), lead standard stock solution (Pb²⁺, 1000 μg/mL, 5 % HNO₃) were purchased from Henan R&D Center for certified reference material. Senkyunolide I (C₁₂H₁₆O₄, Sichuan Naturewill biotechnology Co., ltd, 2203185901), Senkyunolide A (C₁₂H₁₆O₂, Chengdu Purifa technology development Co., ltd, PRF8031301), and Ferulic acid (C₁₀H₁₀O₄, China Institute for identification of pharmaceutical and biological products, 110773–200611) were applied with the purity above 99.0 %. L. chuanxiong Hort samples were purchased from local pharmacies in Chengdu, China.

2.2 Instrument conditions

The analysis of heavy metals was performed by an inductively coupled plasma-optical emission spectrometer (ICP-OES, Agilent 5110). The following wavelengths of the measured elements were used: 188.980 nm (As), 231.604 nm (Ni), 213.857 nm (Zn), 214.439 nm (Cd), 220.353 nm (Pb). Other parameters of the ICP-OES were set as follows: forward power, 1.2 kW; plasma argon flow rate, 12.00 L/min; auxiliary argon flow rate, 1.0 L/min; nebulizer argon flow rate, 0.7 L/min; read time, 10 s; stabilization time; 15 s.

The determination of active ingredients in the decoction of L. chuanxiong Hort was performed on a high-performance liquid chromatography system (HPLC, Thermo UltiMate 3000) connected with an ultraviolet detector. The chromatographic separation was realized by a Thermo BDS C18 column (4.6 mm × 250 mm, 5 μm). The mobile phase consisted of (A) acetonitrile and (B) water (containing 0.5 % phosphoric acid). Analyses were carried out with a gradient elution as follows: 10 %-16 % A at 0–15 min, 16 %-30 % A at 15–30 min, 30 %-40 % A at 30–35 min, 40 %-55 % A at 35–70 min. The detection wavelength was 276 nm. The injection volume was 10 μL. The column temperature and flow rate were set as 30 °C and 1.0 mL/min, respectively.

The in-house fabricated adsorbent was characterized by scanning electron microscopy (SEM, Hitachi SU8020), transmission electron microscopy (TEM, FEI Tecnai G2 F30), Fourier transform infrared spectroscopy (FT-IR, Nicolet IS10), X-ray diffraction (XRD, BRUCKER D8 ADVANCE) patterns, Vibrating Sample Magnetometer (VSM, BKT-4500), and specific surface area and porosity analyzer (Micromeritics ASAP 2460).

2.3 Synthesis of Fe₃O₄@MIL-101(Cr)@PEG

Polydopamine (PDA) has attracted much attention as a functional monomer owing to its unique coating ability and versatility (Duan et al., 2018). It can be self-polymerized onto magnetic nanoparticles under mild conditions to prepare adherent PDA films (Zhang et al., 2012; Zhao et al., 2015). The film with uniform thickness can be formed on various materials through the strong interaction with catechol and amine groups of dopamine(Bai et al., 2017). Therefore, PDA-coated magnetic nanoparticles were prepared as the substrate of MOFs in this study. The synthesis of Fe₃O₄@MIL-101(Cr)@PEG was referenced to our last work with some modification(Tang et al., 2022). Specifically, the synthetic route of magnetic MOF composites was as follows (Fig. 1): firstly, Fe₃O₄ magnetic microspheres were prepared by the solvothermal method according to the reference(Shao et al., 2012). Then, Fe₃O₄ microspheres were coated by a PDA layer via the polymerization of dopamine in the Tris-HCl buffer solution (pH = 8.5). The prepared Fe₃O₄@PDA was dissolved in deionized water containing a certain amount of Cr(NO₃)₃·9H₂O, terephthalic acid, and sodium acetate. The mixture was subsequently sealed in a Teflon-lined bomb and heated under 210 °C for 10 h. In this one-pot synthesis process, Cr(NO₃)₃·9H₂O provided metal nodes while terephthalic acid served as organic building blocks. The obtained Fe₃O₄@MIL-101(Cr) were successively washed with DMF, soaked in hot ethanol for 5 h, washed several times with ethanol, and dried under vacuum at 150 °C. After that, these dehydrated nanoparticles were suspended in PEG solution (1.5 mg/mL in water) and stirred at 30 °C for 4 h. The functionalized nanoparticles were recovered by applying an external magnet and washed twice with water to remove un-grafted PEG. Finally, the adsorbents were washed with ethanol and dried under vacuum at 30 °C.

Close

Fig. 1

Schematic of the preparation of Fe₃O₄@MIL-101(Cr)@PEG and its application for heavy metals removal.

Export to PPT

2.4 Preparation of the decoction of L. chuanxiong Hort

L. chuanxiong Hort samples were randomly collected from local pharmacies. All samples were stored in sealed plastic bags below 4 °C for further use. The decoction of L. chuanxiong Hort was prepared as follows: 2.0 g of L. chuanxiong Hort was weighed accurately into a 250 mL polytetrafluoroethylene crucible. Then, 100 mL of ultrapure water was added to the crucible, boiled for 1 h, and separated from the samples. The last procedure was repeated twice. Finally, the combined decoction was filtered and concentrated under reduced pressure to 0.02 g/mL.

2.5 Adsorption behavior study

Adsorption kinetics, isotherms, and thermodynamic experiments were performed to evaluate the adsorption behavior of Fe₃O₄@MIL-101(Cr)@PEG toward As⁵⁺, Ni²⁺, Zn²⁺, Pb²⁺, and Cd²⁺ (Hayati et al., 2012; Xiong, Xue, et al., 2023). Specifically, the adsorption kinetics were carried out with 40 mg adsorbents in 50 mL of metal ion solutions (50 μg/mL of Pb, Cd, As, Ni, and Zn ions, pH = 5.36). After being shaken for a specific time interval (30–360 min), the kinetic factors were computed. The experimental adsorption data were modeled with the pseudo-first-order and pseudo-second-order kinetic models (Eq. (1) and (2), respectively):

The adsorption isotherms were carried out by adding 40 mg adsorbents into 50 mL of metal ions solution with various concentrations (0.8–100 μg/mL, PH = 5.36). Two well-known Langmuir and Freundlich isotherm models were applied to analyze the isotherm data (Eq. (3) and Eq. (4), respectively):

The thermodynamic adsorption parameters were also measured at different temperature (20 °C, 30 °C, and 40 °C). The mathematical relationship between ΔG, ΔH and ΔS were shown in Eq. (5). The thermodynamic linear relationship was shown in Eq. (6). The values of ΔH and ΔS can be calculated according to the intercept and slope of the fitting curve.

2.6 Removal and analysis of heavy metals in the decoction of L. chuanxiong Hort

As shown in Fig. 1, 40 mg of adsorbents were added to 50 mL of L. chuanxiong Hort decoction. The introduction of magnetism in the material allowed us to separate the adsorbent from the decoction of L. chuanxiong Hort by an external magnetic field. Therefore, the adsorbents were withdrawn by an external magnetic field after being shaken for a specific time interval. Then, the decoction was entirely transferred into a polytetrafluoroethylene crucible and digested with 0.5 mL nitric acid for 2 h. After digestion, the solution was further heated and concentrated to about 1 mL. Finally, the obtained concentrate was entirely transferred to a 10 mL volumetric flask and diluted with 2 % nitric acid solution to volume, which was ready for ICP-OES analysis. To compare the adsorption effect of heavy metal ions, the same experiments were conducted with Fe₃O₄ and MIL-101(Cr), respectively.

2.7 Reusability study

The regeneration and reusability of an adsorbent are imperative for its economic and practical applications. Thus, the adsorption–desorption cycles were repeated for some times to examine the regeneration efficiency of Fe₃O₄@MIL-101(Cr)@PEG. With the method described in section 2.6, the adsorption process was performed by adding 40 mg of adsorbents into 50 mL of spiked decoction with a concentration of 50 μg/mL and an adjusted pH value. After adsorption, the desorption of Pb, Cd, As, Ni, and Zn ions from the adsorbents was carried out with the aqueous HCl solution (0.1 M) as eluent. The Fe₃O₄@MIL-101(Cr)@PEG nanoparticles were stirred in the above solution for 1 h at room temperature. Then, the adsorbents were washed with distilled water several times for the next adsorption–desorption cycle. At the same time, the material after the first desorption was dried at 60 °C and weighed to obtain the recovery rate of the adsorbent.

2.8 Stability study

The stability of Fe₃O₄@MIL-101(Cr)@PEG nanoparticles against acidic solutions is vital for long-time practical applications. Therefore, in the present work, the structural stability of this adsorbent against HCl (pH = 1) was studied for 10 h. Briefly, 100 mg Fe₃O₄@MIL-101(Cr)@PEG nanoparticles were added into 50 mL acidic, and the mixture was stirred at room temperature for 10 h (Ahmadijokani et al., 2021). After that, the nanoparticles were collected, washed with chloroform, and dried at 60 °C. Finally, the XRD diffractograms of the pristine adsorbents, the adsorbents after heavy metal adsorption, and the acid-treated adsorbents were compared to evaluate their structural stability.

2.9 Metal leaching rate

The adsorption process was performed by adding 40 mg of adsorbents into 50 mL of aqueous solution. It should be noted that no heavy metal standard solution was added in this process. After the same operation described in section 2.6, Fe and Cr ions were analyzed by ICP-OES.

2.10 Quality evaluation of the decoction of L. chuanxiong Hort

The content of the representative active ingredients, the solid content, and the HPLC fingerprints similarity were used as the quality monitoring indicators of L. chuanxiong decoction before and after the removal of heavy metals.

3.1 Characterization of Fe₃O₄@MIL-101(Cr)@PEG

The FT-IR spectrum of the in-house fabricated material is shown in Fig. S1. It exhibited the FT-IR spectra of Fe₃O₄, Fe₃O₄@PDA, Fe₃O₄@MIL-101(Cr), Fe₃O₄@MIL-101(Cr)@PEG, and MIL-101(Cr). In the spectrum of Fe₃O₄, the characteristic absorption peak at 606 cm⁻¹ was attributed to the Fe-O-Fe vibration. The two broad bands at 3457 cm⁻¹ and 1640 cm⁻¹ were respectively assigned to O—H stretching and bending vibrations of water, which were probably located on the surface of iron oxide nanoparticles(Shahabadi et al., 2016). In addition to the typical peaks of Fe₃O₄, many new peaks can be seen in the spectrum of Fe₃O₄@PDA. Representatively, the peak at 630 cm⁻¹ was ascribed to the N—H bending vibration. The unique absorption bands near 1290 and 1500 cm⁻¹ were related to the bending vibration of C—N and the stretching vibration of the aromatic ring, respectively. These features showed that the PDA had been successfully coated on the surface of Fe₃O₄. As for MIL-101(Cr), the band at 1021 cm⁻¹ was corresponded to the in-plane bend vibrations of terephthalic acid. The deformation vibration belonging to C—H at 1150, 1013, 889, 751 cm⁻¹ was consistent with the reported peaks of MIL-101(Cr) in literature(Qiu et al., 2020). And as for Fe₃O₄@MIL-101(Cr), the characteristic vibration bands of O—C—O (around 1385 and 1624 cm⁻¹) indicated the existence of the terephthalic acid within MIL-101(Cr) and the connection of the carboxylate ligand to the Cr metal center, which could be obtained in MIL-101(Cr) spectrum(Yalcin & Kayan, 2012). The broad band near 585 cm⁻¹ confirmed the incorporation of the Fe-O-Fe group in the nanocomposite(Madhuri et al., 2005). As for Fe₃O₄@MIL-101(Cr)@PEG, the shifting and increased intensity of the peak in the region 3500–3300 cm⁻¹ indicated the enhanced hydrogen bonding induced by the hydroxyl groups of PEG. The results demonstrated the successful fabrication of Fe₃O₄@MIL-101(Cr)@PEG.

The morphology of Fe₃O₄@MIL-101(Cr) and Fe₃O₄@MIL-101(Cr)@PEG was investigated by SEM and TEM. As described in our last work, the MIL-101(Cr) was verified to be randomly extended around Fe₃O₄@PDA (Fig. S2a and Fig. S2c)(Tang et al., 2022). In comparison to Fe₃O₄@MIL-101(Cr), a rougher surface was obviously observed for Fe₃O₄@MIL-101(Cr)@PEG, which was speculated to be the extension of PEG around the Fe₃O₄@MIL-101(Cr) (Fig. S2b). The TEM image of Fe₃O₄@MIL-101(Cr)@PEG further verified that PEG was loaded around Fe₃O₄@MIL-101(Cr) to construct a core–shell structure (Fig. S2d). The results showed that a larger specific surface area was obtained after the modification of Fe₃O₄@MIL-101(Cr) by PEG. Besides, the overlapped and agglomerated PEG provided a restricted-access structure, which was beneficial for specifically trapping heavy metals.

Since the specific surface area is another important property of adsorbent material, the porous structure and property of Fe₃O₄@MIL-101(Cr)@PEG were tested in this study. Generally, type IV isotherm is generated by mesoporous solids. A typical feature is that the adsorption branch of the isotherm is inconsistent with the desorption branch of the isotherm, and a hysteresis loop can be observed. In this study, a type IV curve with a small hysteresis was provided as the N₂ adsorption–desorption isotherm of the material in Fig. S3, indicating the typical property of a mesoporous structure. The Brunauer-Emmett-Teller (BET) surface area was 248.5447 m²/g, which was satisfactory for enrichment purposes. Moreover, the total pore volume was calculated to be 0.17 cm³/g by the Barrett-Joyner-Halenda method. The average pore diameter of the amorphous nanomaterial was about 4.59 nm. The results have verified that the in-house fabricated Fe₃O₄@MIL-101(Cr)@PEG had a high surface area, large pore volume, and narrow pore diameter. According to the above results, the Fe₃O₄@MIL-101(Cr)@PEG was characterized by beneficial structure and properties for adsorption and separation of heavy metals, which was simultaneously expected to avoid the influence on the active ingredients.

XRD patterns were used to identify the crystal structure of the synthesized nanomaterials(Xiong, Liu, et al., 2023). Fig. S4 exhibited the XRD patterns of MIL-101(Cr) and Fe₃O₄@MIL-101(Cr)@PEG. The sharp diffraction peaks indicated that the MIL-101(Cr) was a material with high crystallinity. The obvious peaks at 2θ of 5.0°, 5.9°, 9.8°, 10.5°, represented the reflections of typical MIL-101(Cr)(Li et al., 2020). After intercalating Fe₃O₄ nanoparticles into MIL-101(Cr), the characteristic peaks of Fe₃O₄ at 30.0°, 35.4°, 54.5°, 63.2°, 67.6° could be observed(Davodi et al., 2019). Obviously, the characteristic diffraction peaks of MIL-101(Cr) were presented in the Fe₃O₄@MIL-101(Cr)@PEG, indicating that the Fe₃O₄ nanoparticles didn’t change the high crystallinity of MIL-101(Cr)(Abutalib & A.Rajeh, 2020). In addition, the peak at 2θ of 19.1° was the characteristic peak of PEG(Zheng et al., 2021), which coincided with the peak of MIL-101(Cr). But it could be confirmed from the intensity of the peak that it was the characteristic peak of PEG. All these characterizations supported the successful preparation of the Fe₃O₄@MIL-101(Cr)@PEG nanocomposites.

The magnetic properties of the Fe₃O₄ and Fe₃O₄@MIL-101(Cr)@PEG were studied using a VSM in an external magnetic field from − 12,000 Gs to + 12,000 Gs. As shown in Fig. S5, the hysteresis loops of Fe₃O₄ and Fe₃O₄@MIL-101(Cr)@PEG were no hysteresis, which showed no remanence or coercivity at room temperature. The results exhibited the typical superparamagnetic character of both materials. The maximal saturation magnetization of Fe₃O₄ was 62.98 emu/g. As for Fe₃O₄@MIL-101(Cr)@PEG, the saturation magnetization decreased to 40.42 emu/g. Although the saturation magnetization was lower than that of bulk magnetite (92 emu/g)(Han et al., 1994), the particles were only redispersed under slight shaking when the magnetic field was removed. This feature facilitated collection, regeneration, and reutilization of the materials. Thus, it could ensure the easy separation of Fe₃O₄@MIL-101(Cr)@PEG from the aqueous solution with a low magnetic field. These structural characteristics and magnetic behavior make Fe₃O₄@MIL-101(Cr)@PEG nanocomposites great potential as a good adsorbent for removing heavy metals from Chinese medicinal materials.

3.2 Removal of heavy metals in the decoction of L. chuanxiong Hort

Parameters that can influence the removal efficiency of the heavy metal ions were initially investigated. All experiments were conducted in triplicate with 50 mL of spiked L. chuanxiong Hort decoction, of which the pH was adjusted with NaOH solution. The spiked concentration of each heavy metal ion was 50 μg/mL, including Pb, Cd, As, Ni, and Zn ions. At the same time, parallel experiments were conducted on L. chuanxiong Hort decoction without spiking heavy metals for the deduction of background.

3.2.1

3.2.1 Effect of pH

The solution pH is a key factor for the adsorption removal of heavy metal ions since it could affect the existing forms of heavy metals and the surface charge of the adsorbents by changing the protonation degree of its functional groups. Therefore, the influence of solution pH on the removal efficiency of Fe₃O₄@MIL-101(Cr)@PEG toward Pb²⁺, Cd²⁺, As⁵⁺, Zn²⁺, and Ni²⁺ is worth studying. In optimization of pH (pH 2-pH 8), the adsorption time and the adsorbent amount were 60 min and 25 mg, respectively. As shown in Fig. 2A, Fe₃O₄@MIL-101(Cr)@PEG exhibited better adsorption performance toward Pb²⁺, Cd²⁺, and As⁵⁺ in a suitable pH range. It displayed that the removal efficiency of Fe₃O₄@MIL-101(Cr)@PEG for all heavy metal ions was lower at low pH and enhanced with increasing pH within a certain range. At the highly acidic solutions, the amine groups and oxygen atoms with lone electrons in the material would be protonated due to the existence of many hydrogen ions. Thus, the positively charged groups on the surface of the adsorbents may repulse the positively charged metal ions. Moreover, H₃O⁺ ions competed with other adsorbate ions for the vacant binding sites on the adsorbents in acidic media. As a consequence, the removal efficiency of all metal ions decreased in conditions of low pH(Kayan & Kayan, 2022). Although the surface charge of the adsorbents may be negative at higher pH, the interactions seemed to be reduced between heavy metal ions and the adsorbents, which might be ascribed to the charge neutralization between the heavy metal and hydroxyl ions. In summary, the removal efficiency increased from pH 2 to pH 6 while slightly decreased from pH 6 to pH 8. It is worth mentioning that the actual pH value of L. chuanxiong Hort decoction is 5.36. The removal efficiency of the heavy metal ions was additionally tested at pH 5.36. As can be seen from the result, the best removal efficiency was obtained at pH 5.36 for Pb²⁺, while pH 6 for As⁵⁺ and Ni²⁺. Besides, the removal efficiency was almost the same for Zn²⁺ and Cd²⁺ at two pH values. Comprehensively considering the removal efficiency and the convenience of practical application, the original pH of the decoction (pH 5.36) was applied in the following experiment.

Close

Fig. 2

Effect of (A) solution pH, (B) adsorption time, and (C) adsorbent amount on the removal efficiency of heavy metals.

Export to PPT

3.2.2

3.2.2 Effect of adsorption time

The adsorption time was evaluated in the range of 30–360 min. In optimization of adsorption time, the solution pH and the adsorbent amount were applied as 5.36 and 25 mg, respectively. Fig. 2B displayed the effect of the adsorption time on the removal efficiency of As, Ni, Zn, Cd, and Pb ions. It can be found that the removal rate was proportional to the adsorption time in a certain adsorption time. The removal efficiency of Zn²⁺, Cd²⁺, and Pb²⁺ almost remained the same from the adsorption time of 240 min, and the removal efficiency of Ni²⁺ and As⁵⁺ reached the equilibrium at the adsorption time of 300 min. Specifically, the adsorption rates followed the sequence: Cd²⁺ ≈ Pb²⁺ > As⁵⁺> Zn²⁺ > Ni²⁺, which could be due to the difference in the hydrated radius, electronegativity, and acid-base properties. To sum up, the best adsorption time was 240 min.

Furthermore, based on Fig. 3 and the data summarized in Table 1, it was evident that the pseudo-second-order kinetic model was more suitable for describing the heavy metals adsorption on Fe₃O₄@MIL-101(Cr)@PEG, suggesting that the adsorption of heavy metals on the Fe₃O₄@MIL-101(Cr)@PEG mainly occurred through chemical adsorption(Xue Zhang et al., 2020). The chemical adsorption probably involved the electrons exchange and sharing (–OH, –NH₃, strong chelating groups), and covalent interactions of active sites with metal ions (As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺)(Xiong et al., 2020; Xiong et al., 2021; Xue Zhang et al., 2020).

Close

Fig. 3

The adsorption kinetics of As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ onto Fe₃O₄@MIL-101(Cr)@PEG (A) and fitted lines with pseudo-first-order (B) and pseudo-second-order (C).

Export to PPT

Table 1 Kinetic factors of Fe₃O₄@MIL-101(Cr)@PEG toward five heavy metal ions.

Kinetic models	Metal ions	R2	Qe (mg/g)	K
Pseudo-first-order	As	0.9835	45.90 ± 1.33	0.0109 ± 0.0010
	Ni	0.9965	26.33 ± 0.37	0.0108 ± 0.0005
	Zn	0.9802	28.27 ± 0.75	0.0128 ± 0.0012
	Pb	0.9808	49.67 ± 1.20	0.0139 ± 0.0012
	Cd	0.9883	46.37 ± 0.96	0.0128 ± 0.0009
Pseudo-second-order	As	0.9974	57.83 ± 1.07	0.0002 ± 0.00001
	Ni	0.9954	33.45 ± 0.87	0.0003 ± 0.00003
	Zn	0.9978	34.61 ± 0.49	0.0004 ± 0.00002
	Pb	0.9989	59.91 ± 0.55	0.0003 ± 0.00001
	Cd	0.9948	56.83 ± 0.33	0.0002 ± 0.00005

3.2.3

3.2.3 Effect of adsorbent amount

The adsorbent amount is another crucial parameter affecting the removal efficiency of heavy metals. The effect of the adsorbent amount was investigated in the range from 10 mg to 60 mg. In optimization of adsorbent amount, the solution pH and the adsorption time were 5.36 and 240 min, respectively. Generally, the ratio of adsorbate ions to adsorbent vacancies was reduced by increasing the adsorbent dosage. Thus, the removal efficiency increased with the adsorbent amount, as shown in Fig. 2C. However, at higher adsorbent dosages, the available vacancies may be reduced due to the aggregation of the adsorbent nanoparticles. The removal efficiency was accordingly predicted to decrease. In this study, the removal efficiency of As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ were close to the maximum when the adsorbent amount was 40 mg. With the adsorbent amount continually increasing from 40 mg to 60 mg, the removal efficiency of heavy metals remained basically unchanged.

In summary, the optimized parameters were obtained as follows: pH 5.36, adsorption time of 240 min, and adsorbent amount of 40 mg. Under the optimal conditions and the spiked concentration of 50 μg/mL, the removal efficiency of Fe₃O₄@MIL-101(Cr)@PEG for As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ were 81.40 %, 51.59 %, 55.40 %, 88.19 %, and 83.79 %, respectively. It indicated that the in-house fabricated adsorbent had excellent removal efficiency for As⁵⁺, Cd²⁺, and Pb²⁺ in the decoction of L. chuanxiong Hort, while obvious removal efficiency for Ni²⁺ and Zn²⁺.

To further assess the removal efficiency of the Fe₃O₄@MIL-101(Cr)@PEG, comparison with Fe₃O₄ and MIL-101(Cr) was carried out under the same conditions. As can be seen in Fig. 4, the Fe₃O₄@MIL-101(Cr)@PEG exhibited the highest removal efficiency for As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺.

Close

Fig. 4

The removal efficiency of Fe₃O₄, MIL-101(Cr), and Fe₃O₄@MIL-101(Cr)@PEG for five heavy metal ions.

Export to PPT

3.2.4

3.2.4 Adsorption isotherms

Adsorption isotherms are critical for the overall improvement of adsorption mechanism pathways and the effective design of an adsorption system. The isotherm data are analyzed using the Langmuir and Freundlich models. The relevant results were depicted in Fig. 5, and the measured parameters were compiled in Table 2. As presented in Fig. 5A, the adsorption capacity of Fe₃O₄@MIL-101(Cr)@PEG toward As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ increased rapidly with an increase in their initial concentrations till the plots reach a plateau and remained relatively constant. The emergence of the plateau signifies that active sites were fully saturated(Ahmadijokani et al., 2021). Fig. 5 unfolded that the Langmuir isotherm model matched the experimental results.

Close

Fig. 5

(A) The adsorption isotherms of As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ onto Fe₃O₄@MIL-101(Cr)@PEG, (B) fitted lines with the Langmuir model, and (C) fitted lines with the Freundlich model.

Export to PPT

Table 2 Isotherm parameters for the adsorption removal of five heavy metal ions on Fe₃O₄@MIL-101(Cr)@PEG.

Isotherm models	Constants	Heavy metal ions
		As	Ni	Zn	Pb	Cd
Langmuir	Qm (mg/g)	60.27 ± 2.68	38.48 ± 1.58	45.06 ± 1.71	72.18 ± 3.28	62.29 ± 0.91
	KL (L/mg)	0.1656 ± 0.0277	0.0758 ± 0.0103	0.0818 ± 0.0103	0.0979 ± 0.0142	0.1442 ± 0.0079
	R2	0.9872	0.9924	0.9935	0.9923	0.9988
	RL	0.057–0.883	0.1166–0.9428	0.1089–0.9386	0.0927-0.9274	0.0649-0.8966
Freundlich	KF (mg/g)	11.8986 ± 3.0280	4.9135 ± 1.2962	5.9556 ± 1.3472	9.8464 ± 2.2955	11.2755 ± 2.3451
	1/nF	0.3940 ± 0.0716	0.4559 ± 0.0686	0.4558 ± 0.0597	0.4763 ± 0.0664	0.4155 ± 0.0587
	R2	0.9119	0.9413	0.9565	0.9494	0.9519

Same conclusion was presented in Table 2. It implied that the adsorption of As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ onto the Fe₃O₄@MIL-101(Cr)@PEG was monolayer. There was no interaction between adsorbate ions, and the surface of the adsorbent was homogeneous(Yuan et al., 2017). As can be seen from Table 2, the calculated R_L and 1/n_F were between 0 and 1, indicating that the adsorption process was favorable and Fe₃O₄@MIL-101(Cr)@PEG was a good adsorbent for heavy metals removal from aqueous solution. This result suggested that the adsorption process of Fe₃O₄@MIL-101(Cr)@PEG toward As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ was monolayer coverage adsorption and favorable(Mahmoodi & Mohammad, 2015; Xue Zhang et al., 2020).

Moreover, the maximum adsorption capacity of the adsorbent toward the heavy metals was compared with the literature data in Table 3. The adsorption capacity of the five heavy metal ions on Fe₃O₄@MIL-101(Cr)@PEG was equal to or better than other adsorbents. Some data showed that the adsorption capacity of this material for heavy metal ions was lower than that of other materials, and it was because those other materials only target one or two heavy metal ions, which greatly reduced the competition for material adsorption sites.

Table 3 Comparison of the maximum adsorption capacity of different adsorbents toward five heavy metal ions.

Adsorbent	Adsorption capacity (mg/g)					Reference
	As5+	Ni2+	Zn2+	Pb2+	Cd2+
UiO-66-NHC(S)NHMe	–	–	–	232	49	(Saleem et al., 2016)
HFB	–	12.13	22.25	2.89	22.03	(Shin, 2017)
Denim textile wastes	1.50	–	25.50	–	14.83	(Mendoza-Castillo et al., 2015)
Granular activated carbon	2.50	–	–	–	–	(Natale et al., 2008)
Fe3O4/Mg-Al-CO3-LDH	–	–	–	–	54.7	(Shan et al., 2015)
Rice husk ash	–	–	14.30	–	–	(Ahmaruzzaman, 2011)
LDH-Cl	–	–	–	108	61	(González et al., 2015)
Modified waste silk	–	7.47	–	–	8.03	(Mia et al., 2022)
Chitosan	–	–	–	7.70	–	(Cheng & Lin, 2006)
Fe2+-loaded activated carbon	2.02	–	–	–	–	(Tuna et al., 2013)
Iron-modified carbon	1.25	–	–	–	–	(Vitela-Rodriguez & Rangel-Mendez, 2013)
Dashukivskij bentonite	–	–	–	52.08	26.04	(Kostenko et al., 2019)
Poultry litter	–	52	80	–	–	(Kucharski et al., 2021)
MgAL-LDH	–	–	–	12	18.25	(Xue Zhang et al., 2020)
Hybrid material-A	–	–	14.29	38.76	–	(Cerraholu et al., 2018)
Hybrid material-B	–	–	25.61	38.09	–	(Cerraholu et al., 2018)
Fe3O4@MIL-101(Cr)@PEG	60.27 ± 2.68	38.48 ± 1.58	45.06 ± 1.71	72.18 ± 3.28	62.29 ± 0.91	This work

The temperature of an adsorption system is associated with adsorption thermodynamics and affects the adsorption efficiency. Thermodynamic adsorption parameters were measured and shown in Table 4. It can be seen that the theoretical maximum adsorption amount (Q_m) of five heavy metals on Fe₃O₄@MIL-101(Cr)@PEG decreased in turn with an increase in temperature from 20 °C to 40 °C, indicating that lowering the temperature was beneficial to the adsorption process. This phenomenon can be explained by the negative values of ΔH of As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺, which implied that the adsorption process was exothermic. Hence, lowering the temperature was conducive to adsorption. As for the ΔG values all less than zero at different temperatures, it indicated that the adsorption process was spontaneous. In addition, the negative values of ΔS indicated that the degree of freedom of the interface between Fe₃O₄@MIL-101(Cr)@PEG and the aqueous solution decreased during the adsorption process. To sum up, the adsorption of As⁵⁺, Ni²⁺, Zn²⁺, Cd²⁺, and Pb²⁺ by Fe₃O₄@MIL-101(Cr)@PEG was a spontaneous, exothermic, and entropy decreasing process.

Table 4 Thermodynamic parameters for the adsorption removal of five heavy metal ions on Fe₃O₄@MIL-101(Cr)@PEG at different temperatures.

Temperature (℃)	Constants	Heavy metal ions
		As	Ni	Zn	Pb	Cd
20	ΔG(kJ·mol−1)	−24.33	−7.11	−8.98	−17.22	−21.89
	ΔH(kJ·mol−1)	−17.65	−2.95	−3.84	−6.02	−14.12
	ΔS(J·mol−1·K−1)	−41.07	−1.18	−2.27	−4.27	−29.93
	Qm (mg/g)	60.27 ± 2.68	38.48 ± 1.58	45.06 ± 1.71	72.18 ± 3.28	62.29 ± 0.91
	R2	–	–	–	–	–
30	ΔG(kJ·mol−1)	−19.75	−7.04	−8.81	−16.43	−18.60
	ΔH(kJ·mol−1)	–	–	–	–	–
	ΔS(J·mol−1·K−1)	–	–	–	–	–
	Qm (mg/g)	55.02 ± 1.34	27.77 ± 0.83	34.76 ± 1.80	62.92 ± 3.10	55.82 ± 2.88
	R2	0.9998	0.9949	0.9995	0.9999	0.9992
40	ΔG(kJ·mol−1)	−16.36	−7.03	−8.68	−15.71	−16.15
	ΔH(kJ·mol−1)	–	–	–	–	–
	ΔS(J·mol−1·K−1)	–	–	–	–	–
	Qm (mg/g)	47.18 ± 1.21	19.62 ± 0.71	26.92 ± 1.35	54.32 ± 2.92	50.19 ± 1.05
	R2	–	–	–	–	–

3.3 Adsorbent reusability

The reusability of the adsorbent was investigated by adsorption–desorption cycle experiments, which was an important parameter to evaluate the applicability of Fe₃O₄@MIL-101(Cr)@PEG. As shown in Fig. 6, the removal efficiency for As, Ni, Zn, Pb, and Cd ions almost remained unchanged after four consecutive adsorption–desorption cycles. The relative standard deviations (RSDs) were calculated to be 5.47 % (As⁵⁺), 8.27 % (Ni²⁺), 5.51 % (Zn²⁺), 7.45 % (Cd²⁺), and 6.60 % (Pb²⁺), respectively. It was demonstrated that the removal efficiency of Fe₃O₄@MIL-101(Cr)@PEG for As, Ni, Zn, Pb, and Cd ions had been well restored at least for four cycles. The result indicated that the in-house fabricated Fe₃O₄@MIL-101(Cr)@PEG had good reusability and a great potential for sustainable heavy metals removal. In addition, the amount of material after the first heavy metal desorption was 34.4 mg. The loss rate was only 14 %, which meant that the recovery rate of the material was 86 %. In this way, only a small amount of adsorbents was added in successive experiment. In general, the recovery rate of the adsorbent was satisfactory.

Close

Fig. 6

Reusability of the in-house fabricated Fe₃O₄@MIL-101(Cr)@PEG.

Export to PPT

3.4 Adsorbent stability

The XRD patterns of Fe₃O₄@MIL-101(Cr)@PEG were shown in Fig. 7. After being used in different solutions (HCl and heavy metals solution) for 10 h, the changes in the crystalline structure of the MOF were insignificant. It was revealed that the Fe₃O₄@MIL-101(Cr)@PEG was quite stable in the acidic solution. Moreover, the material was stable after absorbing heavy metal ions. The obtained results indicated that the synthesized adsorbent had good structural stability and was suitable for the removal of As, Ni, Zn, Cd, and Pb ions from the L. chuanxiong Hort decoction.

Close

Fig. 7

XRD patterns of Fe₃O₄@MIL-101(Cr)@PEG before and after being used in different conditions.

Export to PPT

3.5 Metal leaching rate

The metal ion leaching of the material was tested after use. The resulted leaching rate of Fe ion was 0.00016 % and that of Cr ion was 0.0049 %, indicating that Fe and Cr ions would not cause secondary pollution to L. chuanxiong Hort decoction.

3.6 Influence on L. chuanxiong Hort decoction after heavy metals removal

The quality monitoring indicators of L. chuanxiong Hort decoction were evaluated before and after the removal of heavy metals. In this section, the experiments of heavy metals removal were directly carried out on the L. chuanxiong Hort decoction.

3.6.1

3.6.1 The content of Senkyunolide I, Senkyunolide A, and Ferulic acid

Since Ferulic acid and Senkyunolide are the main effective components of L. chuanxiong Hort, of which the pharmacological effects have been well studied(Lanlan et al., 2022; Zhen et al., 2011), Senkyunolide A, Senkyunolide I, and Ferulic acid were selected as the indicators to evaluate the influence of the adsorbent on the effective components. To evaluate the influence of the adsorption process on the active ingredients of L. chuanxiong Hort decoction, Senkyunolide I, Senkyunolide A, and Ferulic acid were analyzed in the same decoction with and without the removal of heavy metals. The HPLC chromatogram of Senkyunolide I, Senkyunolide A, and Ferulic acid was shown in Fig. 8.

Close

Fig. 8

HPLC chromatogram of (A) the mixed standard and (B) the testing sample.

Export to PPT

The content of Senkyunolide I, Senkyunolide A, and Ferulic acid in the decoction of L. chuanxiong Hort before and after the heavy metals removal was listed in Table 5. Because the change rate was less than 8.00 %, it was clear that the content of the active ingredients was not significantly affected by the adsorption removal process of heavy metals. The in-house fabricated adsorbent exhibited to be able to efficiently remove heavy metals in L. chuanxiong Hort decoction without influencing its active ingredients.

Table 5 Content of the active ingredients in the decoction of L. chuanxiong Hort before and after the heavy metal removal.

Compounds	Quality score (%)						Rate of change (%)
	Control	Sample 1	Sample 2	Sample 3	Sample 4	Sample 5
Ferulic acid	0.1300	0.1189	0.1216	0.1211	0.1211	0.1207	7.18
Senkyunolide I	0.0459	0.0443	0.0456	0.0448	0.0459	0.0447	1.93
Senkyunolide A	0.1283	0.1213	0.1245	0.1234	0.1240	0.1223	4.09

3.6.3

3.6.3 HPLC fingerprints similarity

With the same decoction, one sample solution was directly analyzed by HPLC to acquire the chromatographic fingerprint, while the other five were analyzed after the removal of heavy metals. The similarity evaluation values were all calculated to be greater than 99.9 %. Thus, there was no significant change in the number and area of the chromatographic peaks of L. chuanxiong decoction before and after the removal of heavy metals (Fig. 9). The consistency of the chromatogram profiles indicated that the chemical composition of the decoction remained almost complete before and after treatment. The results further verified that the in-house fabricated adsorbent wound not affect the chemical components of L. chuanxiong decoction during the adsorption removal process of heavy metals.

Close

Fig. 9

Similarity evaluation on HPLC fingerprints of L. chuanxiong decoction with (A_1-5) and without (A) the heavy metals removal.

Export to PPT