Effective Removal of Humic Acid by Zr-MOFs with Surface Modification

Abstract

Humic acid (HA) in makeup water is one of the important safety issues of high-parameter power plants. Herein, the Zr-based metal organic frameworks (Zr-MOFs) were applied to remove humic acid in water. The mesoporous and active sites of Zr-MOFs were controlled by different ratios of ligands to increase the adsorption of HA. The maximum adsorption capacity was 150.15 mg g⁻¹. The morphology and adsorption properties of the Zr-MOFs were characterized using scanning electron microscopy (SEM), X-ray diffraction (XRD), surface charge, Fourier Transform infrared (FT-IR), N₂ adsorption-desorption and adsorption test. The adsorption process of HA accorded with the pseudo-second-order kinetics, while the adsorption isotherm conformed to the Langmuir model and the adsorption was proved to be a spontaneous and endothermic process. Physical adsorption by the mesoporous materials and the hydrogen bonding interactions between the Zr-MOFs and HA were the driving forces of HA adsorption. These results provided useful information for the effective removal of HA and enhanced our understanding of the adsorption mechanism of HA on Zr-MOFs.

1. Introduction

Humic acid (HA) is a kind of supramolecular association of self-assembling heterogeneous and relatively small molecules, which breakdown from the biomass, existing generally in natural water [1]. They are critical hazards when using water, such as the drinking water supply systems and power systems. They could enrich heavy metal or lead to biological instability. Furthermore, HA causes color and taste problems, and can produce disinfection by-products (DBPs), such as trihalomethanes (THMs) and haloacetic acids (HAA) in the drinking water supply system, and with chlorine containing the risk of a carcinogenesis. In power systems, the existence of HA in the makeup water of the power plant may gradually aggregate to the filter surface and cause the water flux of the membranes to rapidly reduce [2,3]. They also inactivate the resin of the ion exchange system [4]. In addition, HA decomposes in a high-temperature and high-pressure environment, and the produced acidic substances cause corrosion on the inner wall of the pipeline in the circulating water system, which affects the safe operation of the power plant [5]. Therefore, it is important to remove HA from drinking and other processed water. At present, different chemical and physical treatments were studied to remove HA, including ion exchange [6], membrane treatment [7], the coagulation sedimentation method [8,9], the oxidation method [10] and photocatalytic degradation [11]. However, these methods still have limitations, such as low removal efficiency. Thus, the question of how to attain high removal efficiency of HA needs further investigation. Adsorption is regarded as one of the most promising methods for removing HA, due to its high efficiency, simplicity and through it being easy to recover and operate [12,13]. Up to now, there many adsorption materials were studied for use in HA removal, such as carbon nanotubes [14], activated carbon [15], graphene [16], resin [17], bentonite [18], kaolin [19], zeolite [20], iron oxide [21], alumina [22] and MOFs [23,24].

Metal-organic frameworks (MOFs) are highly ordered organic–inorganic hybrid materials, formed by the coordination of metal ions and organic ligands [25]. MOFs have the characteristics of a flexibly adjustable pore structure, large porosity and a large specific surface area [26]. Among numerous MOFs, Zr-MOFs show excellent characteristics, such as water, acid, and alkali resistance [27]. In a previous paper, the average size of the HA is 1.6–2 nm [28], which can be removed by adsorption through mesoporous and functionalized matrix materials. However, the structure of Zr-MOFs is mainly microporous, and a large quantity of research has proved that micropores cannot provide an effective surface area for macromolecular HA [29]. The adsorption capacity and adsorption rate are two important directions for its improvement. Therefore, this is an idea worthy of consideration, to seek an adsorbent with a large mesoporous and rich specific surface of the functional group [30]. Therefore, Zr-MOFs adjusted to a suitable structure are expected to enhance the adsorption performance. The pore structure of MOFs can be adjusted by the type of ligand [31], changing the synthesis conditions [32], adding surfactant [33] and so on. Among them, adjustment of the ratio of organic ligands to control the pore structure of MOFs is a promising direction of development. There are some pioneer works in this aspect. For example, Wang et al. [34] introduced various auxiliary ligands to construct three novel Zn-MOFs, and the adsorption of cationic organic dyes was related to the pore size and the charged framework. Michael et al. [35] generated nine Zr-MOFs with different ligands, and the addition of hydrochloric acid accelerated the synthesis without changing the crystallinity and porosity. The surface properties are related to the surface morphology and functional groups of MOFs, thus affecting the adsorption efficiency [36,37]. Zr-MOFs (UiO-66 series), with Zr₆ nodes and ligands such as 1,4-benzenedicarboxylic acid (H₂BDC) and 2-amino-1,4-benzenedicarboxylic acid (H₂BDC-NH₂), were extensively studied in the literature, due to their unusual firm and steady chemical property [38,39].

This study was proposed to increase the adsorption capacity of Zr-MOFs to HA via adjusting the surface properties of the Zr-MOFs with the adjustment of the ratio of organic ligands. For these purposes, the two main purposes of this research were: (a) systematically characterizing Zr-MOFs, such as crystallinity, functional groups, surface topography, pore size distribution and ζ potential and (b) comprehensive analysis of the adsorption process, including the equilibrium, adsorption kinetics, adsorption isotherms and thermodynamics. The findings provided valuable insights into the interaction mechanisms of Zr-MOFs with organic compounds, providing important information, from the effective removal of HA, to the design of Zr-MOFs as adsorbents.

2. Materials and Methods

2.1. Materials

Materials: zirconium tetrachloride (ZrCl₄, 98%, Macleans), terephthalic acid (H₂BDC, 98%, Macleans), 2-aminoterephthalic acid (H₂BDC-NH₂, 95%, Macleans), dimethylformamide (DMF, AR, Shanghai test), ethanol absolute (AR, Zhengzhou). The detailed information about the chemical materials is listed in Table S1, Supplementary Materials. This material was not modified or processed further before use in the experiments. Humic acid sodium salt was acquired from Sigma Aldrich, Berlin, Germany. Humic acid sodium salt (750.0 mg) was dissolved in deionized water (1000 mL) under sonication, and the pH of humic acid solution was adjusted to pH 5.5. Then the solutions were filtered with membrane filters (0.45 μm), and preserved at a low temperature around 4 °C.

2.2. Preparation of Zr-MOFs

This scheme was based on the stable MOF material UiO-66. Different ratios of ligands (the ratios of H₂BDC and NH₂-H₂BDC were 1:0, 1:1, 1:2 and 0:1, respectively) were used to adjust the physicochemical properties of UiO-66, marked as UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66. The specific process is as follows:

Solution A: A total of 5 mmol ZrCl₄ was added to 15 mL DMF and dissolved by ultrasonic over 10 min.

Solution B: A total of 5 mmol of terephthalic acid, 5 mmol of terephthalic acid and amino terephthalic acid (1:1 molar ratio), 5 mmol of terephthalic acid and 2-aminoterephthalic acid (1:2 molar ratio) and 5 mmol of 2-aminoterephthalic acid were added to 15 mL DMF, respectively, and then sonicated over 10 min to dissolve.

The solution B was added to solution A and sonicated for 10 min, and the mixed solution was transferred to the 50 mL autoclave liner. The autoclave was placed in an oven at 80 °C for 12 h, and then at 100 °C for 24 h. The obtained product was placed in a 150 mL beaker, washed alternately in DMF and absolute ethanol three times, centrifuged, and the washed product was vacuum dried at 60 °C for 12 h to obtain Zr-MOFs with different ratios of ligands.

2.3. Characterization

A total of 2 mg of the prepared powder was added to a 1 mL centrifuge tube, then 1 mL of anhydrous ethanol was added, and the sample was sonicated for 10 min to disperse it. A 5 mm × 5 mm conductive adhesive was placed on the carrier table with tweezers and a 5 mm × 5 mm silicon wafer was added, a small drop of the sonicated dispersion was placed on the silicon wafer, and it was placed in an oven at 80 °C for 15 min. The dried samples were put into the gold spraying device together with the carrier table and sprayed with gold for 1 min. The morphology was characterized by a German ZEISS-type field emission scanning electron microscope with the voltage setting of 8–12 kV and the ultimate resolution of 1 nm. The 40 mg sample was placed in the center of the flat slot of the sample table, compacted with a hard plate, scraped off the sample above the upper surface, and the sample was cleaned outside the slot. The X-ray diffractometer is equipped with a CuKα source, a software parameter Twin-secondary of 5, a voltage of 40 kV, a current of 40 mA, a step time of 0.2 s, and a test angle range of 5–40°. A Thermo IS 10 FT-IR spectrometer from the USA was used to characterize the functional groups of the material for FT-IR testing. Dried 2 mg of powder specimens and 200 mg of pure KBr were finely ground under tungsten lamp, placed in a mold and pressed into a transparent sheet on a hydraulic press, and the specimens were put into an IR spectrometer for testing in the wave number range 4000~400 cm⁻¹, scan number 32, and resolution 4 cm⁻¹. For the nitrogen adsorption and desorption test, the specific surface area and pore size distribution of the materials were measured using an ASAP2020 fully automated specific surface and pore size analyzer from Mike’s USA. A 300 mg sample was dried under vacuum at 120 °C for 6 h to remove the water from the pores. The degassing temperature of the instrument was arranged at 120 °C and the degassing time was 6 h. X-ray photoelectron spectroscopy (XPS) was performed using Omnicron XPS.

2.4. HA Adsorption Test of Zr-MOFs

Here, the 0.1 g sample was weighed into HA solution with the magnetic stirrer on. Samples were taken at 0, 0.5, 1, 2, 5, 10, 20, 50, 100, 200, 300, 400, 600, 800, 1000, 1400 and 1800 min, and the absorbance of the solution was detected at 254 nm with an ultraviolet-visible spectrophotometer. To study the process of the adsorption isotherm models, the adsorption tests of HA on the four different Zr-MOFs were recorded at different concentrations (7.5, 15, 30, 45, 60, 90 mg L⁻¹) and different temperatures (15, 25, 35 °C). The results of the removal efficiency of HA were determined by the calibration curve. The following equation was used to calculate the amount of HA adsorbed on the four different Zr-MOFs:

$$\mathit{Adsorption}\mathrm{capacity} (q)=\frac{\left(c_0-c_t\right)v}{m}$$

(1)

Use the following formula to calculate HA removal efficiency:

$$\mathit{Removal}\mathit{efficiency} (R)=\left(\frac{c_0-c_t}{c_0}\right)\times 100\%$$

(2)

Here, q (mg g⁻¹) and R were the adsorption capacity and removal efficiency at equilibrium, respectively. c₀ (mg L⁻¹) and c_t (mg L⁻¹) represented the concentration of HA at time 0 and t h, respectively. m (g) and v (L) were the weight of Zr-MOFs and the volume of humic acid solution.

2.5. Adsorption Kinetics and Isotherms

In order to better comprehend and optimize the adsorption process, the kinetics experiments were investigated using pseudo-first-order, pseudo-second-order models and intra-particle dissemination. These models are described below:

$$lg(q_e-q_t)=lg{q}_e-\frac{k_{1}t}{2.303}$$

(3)

$$\frac{t}{q_t}=\frac{1}{k_2{q}_e}+\frac{t}{q_e}$$

(4)

$$q_t=k_{id}{t}^{1/2}+C_i$$

(5)

Here q_t (mg g⁻¹) and q_e (mg g⁻¹) represented the adsorption capacity of HA on Zr-MOFs at adsorption and equilibrium time, respectively. The above rate constants included k₁, k₂, and k_i are, respectively, from the quasi-first-order, quasi-second-order and intraparticle diffusion models, as shown in Equations (3)–(5).

The two common models of Langmuir and Freundlich were utilized for calculating the adsorption isotherms, as shown in Equations (6) and (7):

$$\frac{C_e}{q_e}=\frac{C_e}{q_m}+\frac{1}{K_L{q}_m}$$

(6)

$$\mathit{ln}{q}_e=\mathit{ln}{K}_F+\frac{1}{n}\mathit{ln}{C}_e$$

(7)

C_e (mg L⁻¹) was the equilibrium concentration of the HA solution; q_e (mg g⁻¹) and q_m (mg g⁻¹) represented, respectively, the equilibrium adsorption capacity and theoretical adsorption capacity; K_L (L mg⁻¹) and K_F (L^1/n mg^−1−1/n g⁻¹) were constant related to Langmuir and Freundlich, respectively. The parameter of heterogeneity was 1/n.

3. Results

3.1. Morphology and Structure of Zr-MOFs

For identification of the synthesized Zr-MOFs, XRD analysis was performed, and the results are shown in Figure 1. Sharp peaks could be observed in the XRD pattern of synthesized Zr-MOFs, which showed they were highly crystalline. The main peaks at 2θ = 7.6°, 8.7°, 12.2°, 17.3°, 22.5°, 25.6° and 31.4° were, respectively, assigned to (001), (002), (022), (004), (115), (224) and (046) crystal planes, which were consistent with the characteristic diffraction pattern of UiO-66. In this figure, the signal peaks of the four different Zr-MOFs showed no significant difference, which indicated that the phase of UiO-66 remained unchanged. However, the diffraction peak positions of NH₂-UiO-66 (1:1) and NH₂-UiO-66 (1:2) were slightly shifted to higher angles in the XRD pattern, which may be caused by the competition of different ligands.

The FT-IR spectra of the obtained materials were used to characterize the functional groups. The spectra of UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66 are shown in Figure 2. The sharp peaks at 1658 cm⁻¹ and 1384 cm⁻¹ were attributed to the stretching of C=O and C=C of Zr-MOFs [40]. Similarly, the peak at 1259 cm⁻¹ is due to the carboxylate (-COOH) of the organic linker in the Zr-MOFs. The peak at 3487 cm⁻¹ was assigned to the symmetric stretching of the -NH₂ bond in the amino-terephthalic acid linker, and 656 cm⁻¹ corresponds to the carbon–carbon double bond of the benzene ring. These signals listed above, through the chemical structure of the Zr-MOFs, indicated that the Zr-MOFs have an abundance of active sites. The FT-IR spectra of UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66 were consistent with the previously reported spectra [41,42]. The N contents of UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66 were found to be 1.68%, 5.46%, 6.62% and 8.23% of N by XPS, respectively. As the proportion of H₂BDC-NH₂ increased, the N elements carried on the ligand increase. The N in UiO-66 probably originated from the organic solvent DMF, that was immersed inside the material during the washing process.

The morphology of four different Zr-MOFs at different ligand ratios were observed through SEM. Figure 3a,b show the original UiO-66, showing a rough granular surface. Figure 3d,f were the SEM images of UiO-66-NH₂ (1:1) and UiO-66-NH₂ (1:2), which, synthesized with the addition of H₂BDC-NH₂, show a more obvious angular shape and uniform ortho-octahedral morphology. The results indicated that the addition of H₂BDC-NH₂ produced better morphology. The average sizes of the four different Zr-MOFs were 90.87 nm, 68.63, 68.41 and 68.16 nm, respectively (Figure 4). The presence of H₂BDC-NH₂ resulted in smaller particle sizes, independent of the amount added. The morphology of the crystalline particles formed was consistent with previous work, which is in agreement with the XRD results.

The N₂ adsorption–desorption isotherms and pore size distribution of the four different Zr-MOFs are shown in Figure 5. The isotherm of the four different Zr-MOFs exhibited a typical type IV adsorption isotherm. There was a hysteresis loop in the high relative pressure area, which reflected the mesoporous in Zr-MOFs. With the increase in the H₂BDC-NH₂, the pore-size of Zr-MOFs increased after the addition of H₂BDC-NH₂. The Brunauer–Emmett–Teller (BET) surface area also increased with the increase in H₂BDC-NH₂. The BET surface areas of UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66 were 498.9, 559.3, 625.5 and 277.1 m² g⁻¹, respectively. This result could be assigned to the increase in microporous and mesoporous with the addition of H₂BDC-NH₂. The relatively low surface area of NH₂-UiO-66 may be due to its fewer micropores. After increasing the ratio of H₂BDC-NH₂, the pore volume of mesopores around 3–10 nm gradually increased. It indicated that the number of mesopores was gradually increasing with addition of H₂BDC-NH₂. The increase in the pore volume and specific surface area in the mesopore range may be beneficial to the adsorption of HA.

At pH 5.5, the Zeta potentials of UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66 were tested to be 13.81 mV, 22.20 mV, 27.09 mV and 31.91 mV, respectively. It was easily seen that, as the H₂BDC-NH₂ ratio increases, the Zeta potential keeps increasing. This was because the amino group of H₂BDC-NH₂ was introduced into the Zr-MOFs as an organic ligand. The Zeta potential of HA, listed in Table S2, Supplementary Materials, was negative at pH 2–11. The Zeta potential of the initial HA solution was tested at −41.64 mV, and the potential value of Zr-MOFs remained positive. This will facilitate the formation of electrostatic interactions to promote the adsorption of the negatively charged HA.

3.2. Kinetics Studies

HA was difficult to remove because of its high solubility and stability in water. However, it is important to improve HA removal due to the public hazard. In terms of the adsorption performance of Zr-MOFs on HA, the adsorption kinetics of Zr-MOFs on HA was studied, and are shown in Figure 6. The adsorption of HA at 25 °C and pH 5.5 increased fast for the first 0.5 h and then gradually reached equilibrium after 10 h. The removal efficiencies were 34.3%, 55.6%, 65.1% and 85.2% for UiO-66, NH₂-UiO-66 (1:1), NH₂-UiO-66 (1:2) and NH₂-UiO-66, respectively. The results exhibited that NH₂-UiO-66 could adsorb HA rapidly with high HA removal efficiency. This change may be significantly influenced by the pore size distribution and active functional groups. With the improved effect of the ratio of H₂BDC-NH₂, the -NH₂ on the surface of Zr-MOFs increased. It was also in line with the increase in the mesopore and the number of functional groups. The qe increased with the ratio of NH₂-H₂BDC, which was consistent with the mesoporous and number of functional groups’ results. These revealed that the surface properties control the Zr-MOFs. For understanding the time dependence of HA removal by Zr-MOFs, kinetic models of the pseudo-first, pseudo-second and intraparticle diffusion were utilized to fit the experimental data. Figure 5b,c showed the results of the adsorption kinetics fitted by pseudo-first-order model and pseudo-second-order model. The correlation coefficients (R²) of the pseudo-second-order models are 0.991–0.999 (

Table 1. Proposed primary and secondary kinetic parameters for the adsorption of HA by Zr-MOFs.

	Pseudo-First-Order		Pseudo-Second-Order
Zr-MOFs	k1/min−1	R2	k2/min · g · mg−1	R2
UiO-66	1.021 × 10−3	0.968	2.475 × 10−4	0.991
UiO-66-NH2 (1:1)	2.30 × 10−4	0.542	6.376 × 10−4	0.999
UiO-66-NH2 (1:2)	3.69 × 10−4	0.683	1.518 × 10−4	0.999
UiO-66-NH2	8.16 × 10−4	0.729	2.365 × 10−4	0.999

), which were higher than that of pseudo-first-order model (0.542–0.968). Therefore, the adsorption kinetics of Zr-MOFs on HA could be better described by a pseudo-second-order kinetic model. In order to investigate the adsorption of different stages of HA removed by four different Zr-MOFs, the fitting curves of the intraparticle diffusion model, which was evaluated for the adsorption process, are shown in Figure 6d. It did not pass through the origin, indicating that intraparticle diffusion was not the only speed-limiting step. The HA diffusion rates of the initial sharp stage (k_i1) at first 1 h (

Table 2. Fitting parameters of intra-particle diffusion model for Zr-MOFs adsorption of HA.

Zr-MOFs	ki1 (mg g−1 min0.5)	ki2 (mg g−1 min0.5)	ki3 (mg g−1 min0.5)	Ci1	Ci2	Ci3	R12	R22	R32
UiO-66	3.088	1.380	0.589	1.888	18.351	38.150	0.989	0.982	0.934
UiO-66-NH2 (1:1)	7.468	1.113	0.151	3.088	58.189	79.958	0.993	0.888	0.978
UiO-66-NH2 (1:2)	7.189	2.223	0.376	6.635	46.888	89.219	0.997	0.918	0.980
UiO-66-NH2	11.232	1.838	0.225	6.028	86.182	122.03	0.995	0.892	0.978

), which was assigned so that the HA migrated from the water to the active sites on the surface of the MOFs. The second stage (k_i2) represents the pore diffusion, where the diffusion rate of the NH₂-UiO-66 was higher than UiO-66. This showed that NH₂-UiO-66 has more available pores and less steric hindrance. The last stage was the equilibrium stage. The HA adsorbed to NH₂-UiO-66 would be dispersed into the water and gradually reach a dynamic equilibrium of HA in adsorbent and dissolved in water. After equilibrium, the internal diffusion was no longer a rate-controlling step.

3.3. Isotherm and Thermodynamics Studies

UiO-66-NH₂ showed excellent adsorption in humic acid solution with a pH of 5.5. Figure 7 showed the adsorption of HA by NH₂-UiO-66 at the different temperatures of 15 °C, 25 °C and 35 °C. As shown in Figure 7a, the adsorption capacity (q_e) increased with the initial concentration increase. The adsorption capacities of NH₂-UiO-66 increased with the increasing adsorption temperature, indicating that increasing the adsorption temperature within a certain range is beneficial to the adsorption. In order to understand the adsorption mechanism of HA by NH₂-UiO-66, the Langmuir and Freundlich models were employed to fit the isotherm parameters, as shown in Figure 6c and Figure 7b.

Table 3. Langmuir and Freundlich model parameters for adsorption of HA on UiO-66-NH₂.

	Model	Parameters	Temperature) (°C)
			15	25	35
NH2- UiO-66	Langmuir	qm/mg · g−1	102.15	140.25	150.15
		KL/L · mg−1	0.302	0.165	0.188
		R2	0.989	0.928	0.973
	Freundlich	KF/L1/n · mg−1−1/n · g−1	42.31	38.55	42.56
		R2	0.744	0.882	0.770

summarizes the corresponding parameters of the two adsorption models. It could be observed that the R² of NH₂-UiO-66 of Langmuir (0.928–0.989) was higher than that of Freundlich (0.744–0.882). These results indicated that the Langmuir model could better describe the adsorption thermodynamics of NH₂-UiO-66 on HA. The isotherm studies illustrated that HA adsorption on NH₂-UiO-66 could be all relevant to the monolayer physical adsorption processes. Mesoporous adsorption and hydrogen interaction lead to physical adsorption. The maximum adsorption capacity (q_m) for HA were also calculated and shown in Table 3. The q_m of NH₂-UiO-66 was 150.15 mg g⁻¹ at 35 °C calculated from the Langmuir model. Previous work reported that the adsorption capacity of HA on other adsorbents was exhibited at between 33.4 and 97.35 mg g⁻¹. The adsorption capacity of NH₂-UiO-66 was 150.15 mg g⁻¹, which was much higher than these adsorbents.

3.4. The Effect of pH on Adsorption

To investigate the effect of pH, the initial pH of the humic acid was adjusted to 2–12 using hydrochloric acid and sodium hydroxide, and the removal efficiencies of humic acid by Zr-MOFs with various ligand ratios were tested. As shown in Figure 8, the removal efficiency of humic acid was significantly higher at lower pH values than at higher pH values. It may be that the carboxyl group binds hydrogen ions at a pH less than 4, and the hydrophobic interaction of the benzene ring and carboxyl group with Zr-MOFs promotes the binding. At pH 4–6, the carboxyl groups start to dissociate, while at a higher pH the phenolic hydroxyl groups start to dissociate and the hydrophobic interactions decrease. At pH 10, the binding mainly relied on the π-π stacking, and when the pH was higher, there was a decomposition of Zr-MOFs, probably due to the hydroxyl radical binding to the carboxyl functional group of Zr-MOFs.

3.5. The Effect of Salt Concentration on Adsorption

To investigate the effect of salt concentration on the adsorption of HA by Zr-MOFs, humic acid solutions with different concentrations of NaCl were prepared, and the removal efficiencies are shown in Figure 9. The removal efficiency of humic acid by Zr-MOFs were all significantly increased with the increase in ionic strength. It was possible that the ionic strength compressed the thickness of the diffuse double layer on HA, which made the Zr-MOFs more accessible to the HA molecules. An increase in ionic strength also minimized the electrostatic repulsion between ionized oxygen-containing groups, which lead to a reduction in the size of HA molecules and a faster diffusion of coiled HA [43].

3.6. The mechanism of Adsorption

According to the results of the characterization and the experiments, when ligands with amino groups were added, the formed Zr-MOFs tended to show higher surface pore sizes with increasing H₂BDC-NH₂. On the one hand, the amino group in H₂BDC-NH₂ exhibited stronger electrostatic interaction, also facilitating the formation of hydrogen bonding to HA. On the other hand, the increase in the pore size was positively correlated with the adsorption capacity, as shown in Figure 10. Thus, the adsorption capacity of HA was increased through active sites and mesopores.

4. Conclusions

The pore size and surface properties of Zr-MOFs were modulated by the addition of two ligands (H₂BDC-NH₂ and H₂BDC) in different ratios (0:1, 1:1, 1:2 and 1:0). The pore size of the mesopores of Zr-MOFs gradually increased, but the crystallinity remained good and the morphology was more regular. NH₂-UiO-66 exhibited excellent HA removal efficiency and the theoretical maximum adsorption capacity could reach 150.15 mg g⁻¹ at pH 5.5. The adsorption process was in accordance with quasi-secondary kinetics and the adsorption isotherms were in accordance with the Langmuir model. The adsorption of HA on Zr-MOFs was found to be driven by the adsorption of mesoporous monolayers on Zr-MOFs. The enhanced effect of humic acid removal from water was due to the increased mesopore size and active sites. With the increase in the number of amino groups and mesopore pore size, the effect of electrostatic interaction and pore size adsorption favors the adsorption of HA. According to this work, the removal of HA in water treatment, by tuning the suitable shape and properties of the ligand-controlled Zr-MOFs, has great potential applications.