Understanding Zinc Transport in Estuarine Environments: Insights from Sediment Composition

Abstract

Sediments are sources and sinks of heavy metals in water, and estuaries are heavily influenced by human production and life. Therefore, it is of great significance to study the composition of estuarine sediments and the relationship between their components to understand the transport and transformation pathways of heavy metals in the environment. In this research, we investigated the characteristics and patterns of Zn adsorption by organic–inorganic composites, organic–clay mineral composites, and iron oxide–clay mineral composites in eight estuarine sediment samples from Dianchi Lake. The results show that both Langmuir and Freundlich isothermal models can describe the adsorption behaviour of the adsorbent better. The order of the adsorption capacity of the three groups of samples for zinc was organic–inorganic composites > organic–clay mineral composites > iron oxide–clay mineral composites. Through FTIR and XRD analyses, the adsorption of Zn²⁺ on the three groups of samples was dominated by electrostatic attraction and coordination adsorption, accompanied by the occurrence of ion exchange and co-precipitation. After FTIR semi-quantitative analysis, it was found that the source of the differences in the high and low Zn adsorption of the three types of samples may be mainly due to the content of phenolic functional groups in the organic matter. This may be related to the low redox site of the phenolic hydroxyl group, which, as an electron donor, is susceptible to electrostatic attraction and complexation with heavy metal cations. The organic–inorganic composite has a higher adsorption capacity for Zn when the ratio of the active fraction of organic matter to the free iron oxide content is 0.65–0.70. In this range, the organic matter can provide enough negative charge without making the sample surface too dense. Iron oxides can also activate the sample by providing sufficient contact between the clay minerals and the organic matter. When this ratio is too high or too low, it will be unfavourable for Zn adsorption.

1. Introduction

The estuary is located between the river and the lake, which is rich in living organisms, energy flow, and material circulation, and is highly susceptible to the superimposed effects of various geological, physical, chemical, and biological factors [1]. In addition, factors such as estuarine runoff and sediment redistribution (input, resuspension, exchange with the surrounding environment, etc.), water salinity and pH, redox, dissolved oxygen, atmospheric deposition, and geological disturbances (including earthquakes) can affect the mobility and spatial distribution of metals through dissolution, deposition, and diffusion [2,3,4]. In the estuary, as the cross section expands, the water flow slows down significantly, resulting in the accumulation of a large number of pollutants in this area. These pollutants are present in various sedimentary forms, and the enrichment of organic matter is one of them [5]. This makes the estuary a hotspot for heavy metal enrichment, and the source and transport of heavy metals is a complex dynamic process [6,7,8]. When heavy metal-rich sediments are disturbed, the zinc in them may be re-released into the overlying water body, and under certain conditions, zinc transport may have a potentially polluting effect on the water body, which is also one of the major pollutants in the water body [2]. Heavy metal contamination of estuarine sediments has been a key research topic in the field of the environment [9]. Therefore, the endogenous pollution caused by the release of heavy metals in sediments has become an environmental problem that cannot be ignored [10,11,12]. Estuarine sediments act as sources and sinks for heavy metals, and usually contain iron oxides, organic matter, and clay minerals that are mostly bound to heavy metals. However, they can be released into the water body with the change or disturbance of the environment at the sediment–water interface, resulting in the secondary pollution of the water body with heavy metals [13,14]. The main process by which heavy metals are immobilised by sediments is adsorption [15]. This pathway directly affects the concentration, transport transformation, bioavailability, solubility, and activity of heavy metals in the environment [16]. The sediments are dominated by a core of various clay minerals (illite, montmorillonite, kaolinite, etc.). Metal oxides (iron, aluminium, manganese oxides, etc.) and organic matter (humic acids, tannins, etc.) form flocculent aggregates on their surfaces [2,17]. Organic matter is bound to mineral surfaces and embedded in the intergranular layers of expansive clay minerals through mechanisms such as hydrogen bonding, ion exchange, deprotonation, hydrophobic forces, anion adsorption, etc. [18]. This affects the tightness of cementation with minerals, as well as colloidal stability, and increases the activation of iron oxides [19]. Strongly reactive hydroxyl groups on the surface of iron and aluminium oxides can bridge with organic matter to produce stable colloids through ligand exchange or the formation of corresponding ionic bonds [20], resulting in the formation of clay minerals as the core, as well as metal oxides and organic matter to participate in the construction of the sediment–organic compound [21,22]. Previous studies have thoroughly investigated the adsorption of pollutants by single components in sediments, but there are fewer studies on the adsorption of heavy metals by composite components relative to the interactions between single components for specific pollutants. This requires an in-depth analysis of the adsorption behaviour of pollutants in each component of the sediment, the use of nonlinear fitting models to obtain the thermodynamic parameters of the adsorption process, and a visual display of the fitting effect to determine the trend in adsorption on pollutants.

Dianchi is the largest freshwater lake on the Yunnan–Guizhou Plateau, located downstream of the main city of Kunming, and is one of China’s key lakes that needs to be managed. It carries the main population and GDP of Kunming, and at the same time, acts as a backup drinking water source—for both industrial and agricultural water—as well as for flood control, aquaculture, and other important functions, subject to the influence of human activities; water pollution is serious. In this research, we remove the organic fraction, the free iron oxide fraction, and the free aluminium oxide fraction from the sediments, respectively, investigating the differences in the Zn²⁺ adsorption characteristics between the composite after the removal of specific components and the original sediment composite. The characterisation differences between the groups of complexes were subsequently observed using scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), and X-ray diffraction (XRD). This study also analyses the adsorption mechanism of Zn²⁺ on natural sediment complexes and explores the interactions and ratios between clay minerals, organic matter, and iron oxides, as well as their effects on Zn²⁺ adsorption [23]. This study will be conducive to the study of the mechanism of fixation and the transfer of heavy metal ions in the sediments of plateau lakes and wetlands, as well as to the adoption of reasonable measures to control the pollution and endogenous release of heavy metals in plateau freshwater lakes according to local conditions [2].

2. Materials and Methods

2.1. Sampling Points

In this experiment, eight rivers entering the lake on the east and south shores of Dianchi Lake in Kunming City, Yunnan Province, were selected for the study (Figure 1). Chai River (CH), Dong-Da River (DDH), Gong-Si River (GSG), Guang-Pu (GPG), Lao-Yu River (LYH), Lao-Bao River (LBH), Xin River (XH), and Yu-Ni River (YNH) were the subjects of the study. Sediment was collected from the top 10 cm of the river bottom using a peat auger at a fixed depth, and each sampling site was located in an estuarine section of the river, where the river flow was slow and the water depth was greater than 0.5 m, so that various types of materials could be more easily deposited and immobilised in the sediments [24].

2.2. Sediment Sample Groups

(1) Without special treatment—Group A: (1) a certain amount of natural sediment complex samples were placed in a beaker, ultrasonically dispersed with the addition of pure water, had their biomass and residues artificially removed, and were freeze-dried to make a product.

(2) Organic matter removal treatment—Group B: Group A samples were freeze-dried by adding H₂O₂ solution (30%) and heating in a 90 °C water bath. This is the precipitate after the removal of the active part of the organic matter.

(3) Iron and aluminium oxide removal treatment—Group C: The samples in group A were added to a sodium citrate solution (1 mol/L) and NaHCO₃ solution (1 mol/L), heated to 80 °C via hydrothermal heating, were then added to Na₂S₂O₄, were washed repeatedly with saturated NaCl solution and purified water, and were freeze-dried.

2.3. Physical and Chemical Property Tests

The sample pH was determined using an acidimeter (OHAUS Starter 3100, Manufactured by OHAUS Instruments Shanghai Co., Located in Shanghai, China). The Total Organic Carbon (TOC), along with the organic matter content (SOM), of the samples was determined with a Total Organic Carbon Analyser (TOC-CRDS, Manufactured by Teledyne Tekmar, Inc. in Monrovia, Calif). The cation exchange capacity (CEC) was determined using a UV spectrophotometer (Agilent Cary 60, Manufactured by Agilent Technologies located in Santa Clara, CA, USA) according to the hexaammonium hexachloride cobalt leaching-spectrophotometric method (HJ 889-2017). The free iron oxide and free aluminium oxide content were determined using ICP-OES (ICPE-9820, Manufactured by Shimadzu Corporation, located in Kyoto, Japan) at wavelengths of 238.204 and 167.081 nm, respectively.

2.4. Isothermal Adsorption Experiment

A total of 0.2 g of sample was weighed into a test tube and 20 mL of 0.01 mol/L NaNO₃ solution was added to avoid precipitation, followed by 10 mL of 2 mg/L, 5 mg/L, 10 mg/L, 30 mg/L, 75 mg/L, and 150 mg/L zinc standard solutions. The pH was adjusted to 6 with NaOH and HNO₃ solutions, and was shaken at 300 RPM/min. After 24 h, the supernatant was filtered through a 0.45 μm membrane and the zinc concentration was determined using ICP-OES. The adsorption volume is calculated using the following formula:

$$Qe={\displaystyle \frac{C_0-C_e}{V}}\ast m$$

(1)

• Qe—adsorption amount, mg/g;
• C₀—initial concentration of Zn in solution, mg/L;
• Ce—concentration of Zn in solution after adsorption equilibrium, mg/L;
• V—volume of solution, L;
• M—sample mass, g.

Langmuir and Freundlich isothermal adsorption models were used to fit the three sediment composites—A, B, and C. The Langmuir isothermal adsorption model describes the homogeneous monolayer adsorption behaviour of the adsorbent on the adsorbent surface. The Freundlich isothermal adsorption model is a semi-empirical equation, which exhibits the non-uniform and multilayered adsorption of the adsorbent on the adsorbent surface [2].

Langmuir isothermal adsorption model:

$${\displaystyle \frac{C_e}{q_e}}={\displaystyle \frac{C_e}{Q_m}}+{\displaystyle \frac{1}{K_L{Q}_m}}$$

(2)

• Qmax—maximum adsorption capacity;
• K_L—adsorption constant.

Freundlich isothermal adsorption model:

$$Ln{q}_e=Ln{K}_F+{\displaystyle \frac{1}{n}}Ln{C}_e$$

(3)

• 1/n—the affinity of the adsorbent to the adsorbent; the smaller the value, the higher the affinity of adsorption;
• K_F—adsorption capacity of the adsorbent to the adsorbent; the larger the value, the stronger the adsorption capacity.

The separation factor R_L was calculated from K_L in the Langmuir isothermal adsorption model and the initial concentration of heavy metals in solution, C₀ (C₀ − 150 mg/L). It was calculated using the following formula:

$$R_L={\displaystyle \frac{1}{1+K_L\ast C_0}}$$

(4)

• R_L—to determine whether it is favourable for the adsorption equilibrium, the smaller its value, the more favourable it is to adsorption
• R_L > 1—unfavourable adsorption
• R_L = 1—linear adsorption
• 0 < R_L < 1—favourable adsorption
• R_L < 0—irreversible adsorption

2.5. Characterisation of Sediment Complexes before and after Zn Adsorption

2.5.1. Scanning Electron Microscope (SEM)

The instrument model is a ZEISS Gemini 300 (Manufactured by Thermo Fisher Scientific in Wilmington, MA, USA); the electron light path is an electron beam without a cross light path in the mirror cylinder; the accelerating voltage is 0.02–30 kV; and the 10 V step is continuously adjustable. The probe beam current is 3 pA–20 nA, the magnification is 12×–2,000,000×, the objective lens is an electromagnetic/electrostatic compound lens, the sample chamber size is 365 mm × 275 mm, and the sample stage travel X = 125 mm; Y = 125 mm; Z = 50 mm; T = −10°–90°; R = 360° (continuously adjustable), with an image acquisition of up to 32 k × 24 k.

2.5.2. Extra-Fourier Spectrometer (FTIR)

The instrument model is a Bruker MPA & Tensor 27 (Manufactured by Bruker, Karlsruhe, Germany) with a wavenumber range of 4000–400 cm⁻¹, a scan count of 32, and a resolution of 4 cm⁻¹.

2.5.3. X-ray Diffraction (XRD)

The instrument model is an Ultimal IV X-ray Cu Ka radiation diffractometer (Manufactured by Rigaku Corporation. Located in Tokyo, Japan) with a wavelength of 0.15406 nm, a voltage of kV, a current of 40 mA, and a scanning range of 10–90°.

2.6. Data Analysis

Excel 2019 was used for collating and performing analyses related to the experimental data; Origin 2021 was used for plotting and the Langmuir and Freundlich isothermal adsorption model fitting; and SPSS 2021 was used for one-way analysis of variance (ANOVA).

3. Results

3.1. Basic Physical and Chemical Properties

As can be seen in

Table 1. Basic physicochemical indicators of natural composite.

Sampling Points	CEC (cmol/kg)			SOM (mg/g)			Free Iron Oxide (mg/g)			Free Aluminium Oxide (mg/g)			pH
	A	B	C	A	B	C	A	B	C	A	B	C	A	B	C
CH	31.27	1.09	9.73	34.39	32.57	44.69	35.51	62.13	7.83	6.06	11.72	1.22	8.49	8.18	9.15
DDH	32.25	0.80	9.63	46.58	20.72	35.89	26.55	41.60	3.37	4.92	4.27	0.81	7.97	7.25	9.5
GSG	45.36	1.48	14.45	49.11	33.66	44.54	30.73	45.71	4.50	4.68	3.42	0.91	8.25	8.14	8.81
GPG	56.87	2.15	26.01	48.81	31.08	66.98	35.23	52.60	5.85	5.33	4.94	1.18	8.02	8.06	8.69
LYH	52.33	1.80	20.92	46.65	21.44	46.90	49.59	76.04	5.79	5.85	8.16	0.81	8.02	7.95	9.18
LBH	20.77	0.811	6.90	37.47	22.66	29.24	28.27	37.03	3.52	4.78	5.89	0.84	7.91	7.39	8.51
XH	18.71	0.44	2.42	31.96	21.52	19.50	8.98	26.52	1.10	1.07	4.71	0.30	8.35	8.39	9.46
YNH	57.92	2.08	19.15	24.83	18.38	24.50	85.05	131.02	17.30	9.59	15.06	2.11	8.06	7.51	9.18

, the high SOM content in group B after the organic matter removal treatment indicates that the natural sediment samples have a low content of active organic matter fractions. The CEC and pH values of group B decreased significantly compared with group A, which is consistent with the trend in the natural soil particles after the reduction in the SOM content [24], and the CEC and pH values of group B decreased significantly [25]. Therefore, we conclude that the oxygenated functional group fractions such as humic acid and fulvic acid were removed from Group B after treatment with H₂O₂ (30%). Group C was derived from clay minerals except for the YNH sample, which may have been incompletely treated due to the high content of free iron oxides, and the other samples, which had very low levels of free iron oxides and favourable alumina, were derived from clay minerals.

3.2. Isothermal Adsorption Model

As shown in

Table 2. Isothermal model parameters for adsorption of Zn by three groups of complexes.

Sampling Points	Parameters of the Freundlich Isothermal Model									Parameters of the Langmuir Isothermal Model
	KF			1/n			R2			Qmax			KL			R2			RL
	A	B	C	A	B	C	A	B	C	A	B	C	A	B	C	A	B	C	A	B	C
CH	6.75	2.12	2.65	0.63	0.85	0.75	0.9993	0.9899	0.9971	27.38	120.2	32.67	0.334	0.014	0.080	0.9994	0.9884	0.9988	0.020	0.323	0.077
DDH	0.69	0.38	0.66	0.52	0.71	0.52	0.9997	0.9998	0.9998	14.16	18.38	9.09	0.012	0.011	0.031	0.9994	0.9998	0.9999	0.357	0.377	0.177
GSG	0.92	0.76	0.80	0.76	0.59	0.40	0.9977	0.9999	0.9999	51.82	15.05	4.63	0.010	0.021	0.117	0.9963	0.9996	0.9998	0.4	0.241	0.054
GPG	1.03	0.17	0.44	0.67	0.86	0.73	0.9978	0.9998	0.9994	22.14	69.05	35.18	0.029	0.001	0.006	0.9981	0.9998	0.9991	0.187	0.870	0.526
LYH	2.08	1.02	1.08	0.18	0.54	0.46	0.9999	0.9988	0.9998	4.51	14.04	8.85	0.370	0.033	0.066	0.9999	0.9989	0.9997	0.018	0.168	0.092
LBH	0.69	0.36	0.51	0.40	0.63	0.66	1.0000	1.0000	0.9999	5.35	12.14	17.60	0.041	0.011	0.013	0.9998	0.9999	0.9998	0.140	0.377	0.339
XH	0.26	0.62	0.43	0.81	0.45	0.59	0.9999	0.9996	0.9998	28.49	6.30	9.92	0.005	0.032	0.018	0.9999	0.9996	0.9999	0.571	0.172	0.270
YNH	1.34	0.50	0.48	0.30	0.67	0.71	0.9986	0.9998	0.9080	5.36	22.74	14.01	0.092	0.009	0.021	0.9123	0.9995	0.9995	0.067	0.426	0.241

, the adsorption process of Zn on the samples of the eight rivers in groups A, B, and C satisfied the Langmuir and Freundlich isothermal adsorption model (R² > 0.908). Their Langmuir and Freundlich isothermal adsorption model parameters are expressed as follows: The R_L values were B > C > A and R_L < 1, which implies that the surfaces of all three complexes are favourable for adsorption, with the degree of favouritism being A > C > B. The 1/n values were B > C > A (0.18–0.86), but most of them were higher than 0.5, except for the following samples: LYH (1/n = 0.18), LBH (0.40), and YNH (1/n = 0.3) in group A; XH (1/n = 0.45) in group B; and GSG (1/n = 0.40) and LYH (1/n = 0.46) in group C had 1/n < 0.5. This means that the Zn adsorption affinity inside the samples is low and most of the adsorption of Zn²⁺ by the samples may occur in the surface layer. Also, the adsorption affinity of Zn inside the three groups of samples showed A > C > B. The K_F value can reflect the adsorption capacity of the samples in multi-component layer adsorption, and most of the river samples showed A > C > B, which is consistent with the trend in the 1/n values. This indicates that the adsorption affinity and adsorption capacity of Zn²⁺ within the samples of groups B and C generally decreased after treatment. Although the performance of some river samples did not match the overall trend, taken together, the adsorption capacity of Zn²⁺ on the three groups of samples, whether monolayer homogeneous adsorption or multilayer non-homogeneous adsorption, showed A > C > B.

3.3. SEM

The electron micrographs of three sets of samples, A, B and C, were selected for comparative observation, as shown in Figure 2. All three sets of samples showed significant lamellar structures, which helped to expose more adsorption sites with a similar but not identical morphology. The surface of Group A samples has broken and scattered flaky structures, which are stacked but not dense, and are made up of organic matter, iron and aluminium oxides, and clay minerals combined and interacting with each other. The surface morphology of the Group B samples was similar to that of the Group A samples, but the interlayer structure was more fragmented, with more and larger pore structures formed due to the removal of active components from the organic matter. Group C samples have a dense lamellar structure, with almost no broken lamellar particles on the surface, and the organic matter encapsulates the clay mineral surface to form a bumpy but dense structure.

3.4. FT-IR

The Fourier transform infrared spectra of the sediment complex before and after Zn adsorption are shown in Figure 3.

A comparative analysis of the Fourier transform infrared (FTIR) spectra before and after Zn adsorption on the sediment complex of Figure 3 was conducred. After adsorption, the intensity of the characteristic peaks of organic functional groups at 3697 cm⁻¹, 3622 cm⁻¹, 3446 cm⁻¹, 2360 cm⁻¹, 1436 cm⁻¹, 920 cm⁻¹, etc., changed and their positions migrated. This suggests that this class of organic functional groups complexes or coordinates with Zn to form a -COO-Zn covalent structure via hydrogen bonding [26]. Zn may also enter the clay mineral interlayers and replace the Si-O-Mg covalent structure via an ion exchange mechanism to form Si-O-Zn covalent bonds [27]. Carbonate minerals, on the other hand, can precipitate Zn via an electron transfer mechanism [28]. The stronger characteristic peaks of alcohols, phenols, and carboxylic acids in the Group A samples can effectively enhance the activity of the samples and their complexation and coordination with Zn, creating more high-energy binding sites for the samples [29]. The sudden decrease in the number of silicide functional groups in Group B samples may originate from the disruption of the organic matter envelope, leading to a decrease in clay mineral activity. Group C samples are similar to Group A, with the presence of a large number of hydroxyl and carboxyl functional groups, which may be related to the formation of ionic and hydrogen bonds and coordination interactions [26].

3.5. XRD

X-ray diffractograms before and after Zn adsorption on the sediment complex are shown in Figure 4.

The variation in the X-ray diffraction characteristic peaks before and after Zn adsorption in the sediment complex can be seen in Figure 4. The variability of SiO₂ diffraction peaks in the three groups of samples may be due to the masking effect of Zn after adsorption. It has been shown that the adsorption of heavy metals by unmodified SiO₂ is extremely weak [30]. The overall weakening or disappearance of the diffraction peaks of carbonate minerals (calcite and dolomite predominate) may be due to the high adsorption affinity of carbonates for heavy metals. Zn is a soft Lewis acid, easily coordinated to the corresponding Lewis soft base atom [31]. Between layers of carbonate minerals, Zn can be exchanged with Ca, Mg, and other ions therein [27], whereas the higher pH of the sample favoured the formation of Zn-carbonate precipitation [32]. The above is consistent with the phenomena observed using FT-IR. The overall weakening or disappearance of the characteristic peaks of iron oxides may be due to their interactions with organic matter and clay minerals. The hydroxylated surface of iron oxides can adsorb oxygenated acid radicals exclusively, coordinate with oxygenated functional groups on the surface of organic matter to form a covalent structure (Fe-O-C) or form a cationic bridge, releasing OH- and H₂O, thus adsorbing Zn [33]. In addition, the tight binding of iron oxides to silicates can lead to changes in their adsorption capacity and play a role in the adsorption process, no less than that of clay minerals [34,35,36]. Mica-like minerals are electronegative and can be electrostatically attracted to Zn. The Si-OH of seafoam can be surface complexed with Zn, and Zn can be immobilised via homocrystalline substitution into its interior.

4. Discussions

It has long been shown that humic substances are more capable of adsorbing metals than inorganic mineral components [37]. As such, Zn²⁺ can be directly fixed by organic matter complexation [38]. Organic matter is rich in negative oxygen functional groups, which can reduce the zeta potential of the sediment complex and enhance the electronegativity, facilitating its electrostatic attraction with Zn²⁺. However, after the loss of electrons, the morphology of the organic matter changes and the looser structure makes it more hydrophobic [39,40].

To further explore the role played by organic matter in the three sets of samples, semi-quantitative analyses in FTIR were performed. As can be seen in Figure 5, the peak areas of these functional groups before and after adsorption were significantly different, and the peak areas were all reduced after adsorption. The adsorption of zinc by the three sediment complexes is not confined to the molecular layer of organic matter on the surface, but goes deep into the interior of the complexes. The spectral features in the mid-infrared region (4000–400 cm⁻¹) of groups A and C indicate that alcohols, phenols, and carboxylic acids are involved in the adsorption process by directly complexing with Zn [41]. The magnitude of the peak area change after adsorption was mostly smaller in group B than in groups A and C, which indicated a decrease in the adsorption capacity of its surface oxygen-containing functional groups with Zn.

In order to identify the oxygen-containing functional groups that have a decisive influence on the adsorption process of the three groups of samples, the absorption peak areas of different molecules were used to roughly quantify the compounds in conjunction with the semi-quantitative analysis using FTIR, and the degree of change in the peak area before and after the adsorption of the three groups of samples was analysed using ANOVA; the results are shown in

Table 3. ANOVA analysis of FT-IR characteristic peak area changes before and after Zn adsorption in three groups of sediment complexes.

Wavelength	F	P (0.05)	Significance
920 cm−1	0.064	0.938	insignificant
1436 cm−1	4.841	0.019	significant
3446 cm−1	3.819	0.038	significant
3622 cm−1	1.0160	0.379	insignificant
3697 cm−1	3.191	0.062	insignificant

. The phenolic hydroxyl group showed a significant difference in peak area before and after adsorption in the three groups of samples, there was no significant difference in the peak area of carboxyl group and alcohols, and phenols played a dominant role in the adsorption process of Zn. In general, high concentrations of organic matter bound to the mineral surface inhibit the entry of Zn. However, in the solution system, the proton substitution mechanism with the intervention of water molecules can adsorb Zn²⁺ to the sample surface [42]. The phenolic hydroxyl group has a lower redox potential and is often preferentially used as an electron donor in redox reactions; thus, the negative charge it carries can be electrostatically attracted with Zn²⁺ [43]. At the same time, as a polar oxygen-containing functional group, it is more hydrophobic and is more susceptible to the complexation of polybasic ligands with the central ion (Zn²⁺) to form cyclic chelates [44,45]. Although the change in peak area of other oxygen-containing functional groups is not significant before and after adsorption, such functional groups form an organic film on the surface of clay minerals, which improves the indicated activity and enhances the adsorption of iron oxides with clay minerals [29]. The stretching vibrations of silicides (Si-O-Mg), siloxanes (Si-O-Si), and alkyl chains (-CH) in the infrared spectra may be relevant.

Iron oxides also play an important role in the adsorption of heavy metals. Iron oxides possess abundant iron hydroxyl (Fe-OH) groups on their surfaces, which have a high immobilisation capacity for metal ions [34,35]. The presence of iron oxides increases the specific surface area of clay minerals [46]. The binding between it and the silicate promotes the adsorption of Zn, while its own Zn adsorption capacity is no less than that of clay minerals [34,35,36]. It has now been found that clay minerals coated with iron oxide have an enhanced metal-bearing capacity [35], leading to the formation of zinc complexes on the surface of acicular ferrite-coated kaolinite [47]. According to Hochella et al. [48], heavy metals generated from human production and living activities have entered the crystal structure of clay minerals. And Péter Sipos et al. [36] further found that mineral particles covered with iron oxide increase the metal-bearing capacity of neighbouring minerals.

As seen in Figure 5, group B has a smaller phenolic functional group peak area compared to group A. This may imply that the organic matter removed during H₂O₂ treatment may be the more reactive fractions represented by phenols. In order to explore the relationship between iron oxides and the active fraction of this fraction of organic matter in more depth, the content of iron oxides in group B was referenced to the content of organic matter in group C. The content of the active fraction of organic matter in the samples of group A was estimated using proportional conversion, resulting in

Table 4. Content of organoplasmic active fractions and their ratio to free iron oxide content in samples of group A.

Sampling Points	Organic Matter Active Component (mg/g)	Organoplasmic Active Fraction/Free Iron Oxide
CH	15.77	0.4442
DDH	33.36	1.2563
GSG	26.48	0.8617
GPG	27.99	0.7946
LYH	32.67	0.6588
LBH	20.17	0.7135
XH	24.67	2.7476
YNH	12.90	0.1517

.

A comparison of Table 2 with Table 4 reveals that there is no significant pattern between phenolic content and the adsorption capacity of the samples for Zn, whereas the organoleptic active fraction/free iron oxide ratio shows a clear trend. The ratio of organic active fraction/free iron oxide was about 0.65–0.7, which corresponded to the strong Zn adsorption capacity of the samples. This ratio is either too high or too low for the complex to adsorb zinc. Within this range, the organics provide a sufficient negative charge without making the sample surface too dense. YNH had the lowest ratio of organic active fraction/free iron oxide, but its adsorption effect was not the worst. XH had the highest ratio and the worst Zn adsorption performance. The samples with the highest ratio of organic active fraction/free iron oxide had the lowest Zn adsorption performance. The adsorption capacity of the samples with organoplasmic active fraction/free iron oxide ratios between YNH and LBH did not differ much, and LYH had the strongest adsorption affinity among them, after which the adsorption capacity decreased significantly with the increase in the ratio. Therefore, we suggest that the main driver of Zn adsorption by natural complexes is also the ratio of organoleptic active fraction/free iron oxide. Based on the performance of the YNH samples, it is possible that one or more intervals of high adsorption capacity for heavy metals such as Zn may exist at lower values of the ratio of organoplasmic active fraction/free iron oxide.

The adsorption of Zn²⁺ by clay minerals may also play an important role after Zn²⁺ enters the interior of the complex. Among them, carbonate minerals have a high adsorption affinity for heavy metals such as Zn²⁺, with which carbonate precipitation, co-precipitation, and ion exchange occur [32]. Zn²⁺ enters the interlayers of silicate minerals, with which it undergoes coordination. In summary, the adsorption of Zn on group A samples mainly relies on the electrostatic attraction and coordination between Zn and phenolic functional groups. As shown in Figure 6, the adsorption process may also be accompanied by ion exchange, co-precipitation, and so on.

The natural sediment complex is a complex, and arbitrarily small differences between components can lead to large differences in the adsorption results of heavy metal ions by the complex. Each of the eight streams is located in a different clay mineral distribution area, with relatively high kaolinite content (mean 45.6 percent) within the GPG, GSG, and LHB distribution areas; there is a high illite content (mean 41 percent) in the CH, DDH, LYH, XH, and YNH distribution areas. Differences in the major constituents of clay minerals can lead to dramatic differences in the adsorption of Zn²⁺ by sediment complexes. Whereas the combination of minerals is variable, certain combinations of minerals may result in the significant immobilisation of metal ions over a wide range, or may result in a very low adsorption capacity [36].

During human and natural activities, various types of heavy metals can reside in the octahedral structure of clay minerals, along with iron oxides, to increase load bearing [49]. In contrast, such structures in the clay minerals of groups B and C, after the removal of iron oxides and organic matter fractions, may be disrupted, leading to large changes in their Zn²⁺ adsorption capacity.

Changes in the state of the organic matter envelope of the complex are caused by excessively high or low ratios of phenolics/free iron oxide. The loose organic matter layer is undoubtedly more conducive to the adsorption of Zn²⁺, and when the organic matter content is too high, it will form a thicker encapsulation layer on the surface of the composite, blocking the heavy metal pollutants on the surface of the composite. During the removal of SOM by H₂O₂, sediment particles in the form of microagglomerates are dispersed and the properties of the sediment complex particles are changed. The decomposition of inorganic materials may have occurred during this process, forming new mineral species and increasing the content of metal oxides, affecting the particle size distribution of the sediment complexes and ultimately having an impact on the adsorption results of the Group C complexes [49,50,51].

It was shown that the total pore volume and porosity of the sediment complex decreased significantly after the removal of organic matter and iron oxides, especially after the removal of iron oxides [52]. And this may be an important factor contributing to the weakening of Zn²⁺ adsorption by the B and C complexes, combined with Freundlich isothermal adsorption model parameter guessing. Group B complexes lose the reticulated organic matter envelope that is rich in negative oxygen functional groups, the negative charge plummets, and the activity of iron oxides and silicate clay minerals decreases. At the same time, its pH decreases, which is unfavourable for the formation of metal carbonate precipitates in the complex, thus showing the worst adsorption capacity for Zn²⁺. Group C complexes were removed due to the loss of iron oxides, and the organic matter envelope was tightly adhered to the complexes, making it difficult for Zn²⁺ to enter the interior of the complexes to be adsorbed by the clay minerals and iron oxides. However, the organic matter–clay mineral composite, as a whole, carries more negative charge and negative oxygen functional groups, which may still adsorb Zn²⁺ through hydrogen bonding, electrostatic attraction, coordination, and complexation.

5. Conclusions

The results showed that both Langmuir and Freundlich isothermal models could describe the adsorption behaviour of the adsorbent better. The adsorption capacities of the three sediment complexes for Zn were A > C > B, i.e., organic–inorganic composite > organic–clay mineral composite > iron oxide–clay mineral composite. The main reason for the different adsorption capacities of the three complexes on Zn may be affected by phenolic organics. The adsorption of Zn²⁺ by the group A complexes was dominated by electrostatic attraction and coordination adsorption, accompanied by ion exchange and co-precipitation, while ion exchange and co-precipitation occurred at the same time. When the ratio of the active fraction of organic matter to the free iron oxide content was 0.65–0.70, the organic–inorganic complexes had a strong adsorption capacity for Zn.