Characteristics of Cu(II) and Zn(II) adsorption on sediments from typical urban polluted rivers in Dianchi Lake


 This study was conducted to determine the spatial distribution characteristics of Cu and Zn adsorption on the sediments of the estuary of Dianchi Lake, as well as the adsorption laws of Cu and Zn on combinations of sediment organic matter, metal oxides, and organic-inorganic composites.A static adsorption experiment was applied to four groups of sediments from the estuary of Dianchi Lake, and results were generated through correlation analysis and redundancy analysis. The four groups were as follows: (1) Untreated, Group A, (2) Organic matter removed, Group B, (3) Iron and aluminium oxide removed, Group C, (4) Organic matter and iron and aluminium oxide removed, Group D. The adsorption capacity was correlated with the spatial distribution along the direction of river flow and vertical depth. High contents of various components of the sediment did not correlate with high adsorption capacities for Cu and Zn, according to the use of four groups of sediments subjected to different treatment processes. The adsorption of Cu fit the Freundlich isotherm adsorption model for all four sediment groups. For Zn adsorption, the untreated and removed organic matter and Fe-Al oxide groups were in good agreement with the Freundlich model, while the removed organic matter and removed Fe-Al oxide groups were in good agreement with the Langmuir isothermal adsorption model. The results indicate that there is a quantitative relationship between the adsorption of heavy metals and organic and inorganic complexes in sediments.


Introduction
Energy ow and material circulation in estuaries, the transitional areas between lakes and rivers, are strong. Estuarine circulation, which results in river sediment redistribution, water salinity, redox processes, and pH, affects the mobility (including dissolution, deposition, and diffusion) and spatial distribution of metals ( 1 ). The amount of heavy metals entering Dianchi Lake varies greatly with the seasons due to limited surface runoff during the arid dry season and considerably higher surface runoff in the heavy rains of the rainy season. In addition, as the urbanization of the rivers owing into Dianchi Lake continues to increase year by year, pollution sources are also increasing. The large amounts of heavy metals carried into Dianchi Lake readily accumulate in the estuary 2 . Previous studies on heavy metals in Dianchi Lake focusing on the spatial distribution, morphological analysis and ecological risk assessment of heavy metals 3 are mostly based on the direction of adsorption of heavy metals on the sediments.
Adsorption is the dominant process of pollutant transfer to sediments 4,5 , directly affecting the migration, transformation, bioavailability, solubility, and activity of heavy metals in the environment, as well as the concentration and bioavailability of heavy metal ions in sediments 6 . Therefore, exploring the adsorption mechanism and distribution characteristics of heavy metals in estuarine sediments is conducive to formulating corresponding prevention and control measures according to local conditions, as well as to improving the understanding of the ltering and puri cation of heavy metal pollutants in wetland ecosystems and their role in secondary pollution 7 .
Natural water sediments are mainly composed of minerals, with clay minerals (illite, kaolinite, and montmorillonite) as the core substances. Metal oxides (iron oxide, alumina, and manganese oxide) and organic matter (OM, mainly humic acid and tannin acid) bind to the surface of core mineral particles and form occulent aggregates, which play an indispensable role in the adsorption process of metal pollution 8,9 . Therefore, relevant scholars have performed much research on the role of sediment components in the process of adsorbing heavy metals. Typically, the role of single components in adsorption is studied by arti cially synthesizing single sediment components 10,11 or by selectively removing certain components (e.g., OM or iron oxide) from the sediment 11,12 . However, the study of single components overlooks the in uence of organic-inorganic composites in multi-component sediment 13 . It is necessary to consider the role of each potentially in uential component to comprehensively characterize sediment adsorption. In this study, we endeavoured to further explore the adsorption mechanism of multicomponent interactions and improve the understanding of the multi-component complex adsorption system theory.
In this study, we investigated the effects of spatial distribution characteristics on the adsorption capacities of copper (II) (Cu) and zinc(II) (Zn) in sediments at various vertical depths in the estuary and at various distances from the river in the direction of ow. We also investigated the in uence of OM, inorganic metal oxides, and organic/inorganic complexes on the adsorption of Cu and Zn to further understand the settling rule of heavy metals in natural water bodies, to improve comprehension of the Dianchi Lake environment, and to provide technical support for heavy metal pollution prevention and control.

Materials And Methods
Sediment sampling. The sampling points ( Fig. 1) are located in the Xin River next to the Positioning Observation Station of the National Plateau Wetland Research Center. Sediment samples were collected using a xed-depth peat drill at 10-cm intervals along the direction of water ow. Three depths were sampled at each interval: (1) 0-10 cm, (2) 10-20 cm, and (3) 20-30 cm. After removal of impurities, airdrying and grinding, samples were passed through a 0.2-mm sieve and stored in sealed Ziplock bags.
The pH of the sediment was measured by the glass electrode method with a water-soil ratio of 1:2.5. Sediment OM was obtained by using a total organic carbon (TOC) analyser to obtain the TOC content and then converting this value with a coe cient. The cation exchange capacity (CEC) was measured by hexaammine cobalt trichloride extraction spectrophotometry. The contents of iron and aluminium oxide were determined by extraction with oxalate-ammonium oxalate solution. Table 1 shows the main physicochemical properties of the studied sediments.  Data analysis and statistical treatment. The Freundlich (1) and Langmuir (2) isothermal adsorption models were used to t the experimental data: where Q is the equilibrium adsorption concentration (mg/g), Ce is the equilibrium concentration (mg/L), K f is the equilibrium adsorption coe cient, and 1/n is the linearity degree of the isothermal adsorption curve: 2 where Q is the equilibrium adsorption concentration (mg/g), Qe is the maximum adsorption capacity (mg/g), K l is the equilibrium adsorption coe cient, and Ce is the equilibrium concentration (mg/L).
Excel 2013 and SPSS 20.0 were used to collate and perform the correlation analysis of the experimental data, Origin 2018 software was used for chart analysis and isothermal adsorption model tting, and Canoco 5 was used for redundancy analysis.

Results
Adsorption capacity and distribution characteristics of Cu and Zn in estuarine sediments. Along the ow of the river to the estuary, changes in sedimentation of both Cu and Zn were not signi cant at 0-10 cm depth, according to the results of the absorption studies. At 10-20 cm, both Cu and Zn absorption decreased. At 20-30 cm, Cu adsorption gradually increased, while Zn adsorption gradually decreased with distance from the river (Fig. 2).
The content of each component of the sediment was linearly tted with the distance from the river (Table 2), and the slope of the tted line was used to characterize the change trend of each component, with a positive slope indicating an increase in concentration and a negative slope indicating a decrease. At 0-10 cm, OM and CEC increased with distance from the river, while iron and aluminium oxides decreased. At 10-20 cm and 20-30 cm, all components, except for CEC, decreased with increasing distance from the river.
The vertical distribution of the various sediment components showed that the content of each component was greater in the middle sediments than in the surface and deep layers. Figure 3 shows that the OM, CEC, and iron-aluminium oxide contents of the sediments at a depth of 10-20 cm were higher than those at the other two depths. Statistical analysis of isothermal adsorption. As shown in Fig. 4 and Table 3, the difference in the adsorption of Cu in group A was much greater than that in the other three groups. The adsorption of Cu in group A increased to 1.54 mg/g, while the highest adsorption capacities of groups B, C, and D were 0.17, 0.22, and 0.18 mg/g, respectively. The isotherm adsorption curve gradually stabilized, and the adsorption process tended to reach a balance. Adsorption equilibrium was not reached for Group A, the untreated group.
The correlation coe cients (R 2 ) indicate that all four groups showed a good t with the Freundlich isothermal adsorption model, indicating that the adsorption of Cu was dominated by multi-layer heterogeneous adsorption. The adsorption capacity of the four samples followed the order A > C > B > D, and the order of the K value (representing the adsorption force in the Freundlich model) was also A (6.72) > C (0.17) > B (0.13) > D (0.1).
The fact that 1/n > 1 (1.02) in the isotherm adsorption curve of Group D indicates that Group D follows a linear adsorption process and that the adsorption process is di cult to carry out. Additionally, given that 1/n > 1 (1.29) for Group A, it is possible that Group A is far from reaching adsorption equilibrium, making the adsorption isotherm curve present a similar linear adsorption process. For the other two groups, 1/n < 1 (0.47 and 0.76), indicating that the adsorption process is nonlinear adsorption.
As shown in Fig. 4 and Table 4, the isothermal adsorption curves of Groups A, B, C and D gradually stabilized for Zn, and the adsorption process tended to reach equilibrium. The Freundlich model provided consistent results for Group A and Group D, indicating that their adsorption process was dominated by multilayer heterogeneous-phase adsorption, while Group B and Group C were more in line with the Langmuir isothermal adsorption model. The four groups of samples presented values of 1/n < 1 (0.04, 0.03, 0.03, 0.63) in the Freundlich model, indicating that the adsorption process was non-linear. Nevertheless, the adsorption results showed similar trends for Zn and Cu, with the K value following the order Groups A > C > B > D.

Discussion
Correlation analysis between the content of each component and the adsorption capacity of the sediment in the direction of water ow was carried out. The correlation between the adsorption amount and the OM content, CEC, and iron and aluminium oxide contents did not correspond to the change trend of the adsorption amount. It can be seen from  To further explore the relationship between OM content, CEC, iron and aluminium oxide contents and adsorption capacity in the direction of water ow, redundancy analysis (RDA) was applied using Cu and Zn as species and the OM content, CEC, and iron-aluminium oxide content in the sediment as environmental factors. The analysis results showed that the correlation between the adsorption of Cu and Zn and each component of the sediment was different. Figure 5 shows Cu, iron oxide and aluminium oxide at angles less than 90°, indicating that iron and aluminium were correlated more strongly with Cu adsorption than were other components of the sediment, while OM may be more strongly correlated with Zn adsorption.
Tables 6 and 7 show that the contribution rates of OM and aluminium oxide in the adsorption process were higher than those of CEC and iron oxide. The contribution rate of OM in the entire adsorption process reached 72%, with a maximum signi cant difference of 0.002. Based on the distribution characteristics of the adsorption amount with increasing distance from the river and the redundancy analysis results for the sediment components and adsorption amounts, the adsorption of Cu in the Dianchi Lake Estuary is strongly affected by iron and aluminium oxides, while the adsorption of Zn is more affected by OM and CEC.  17 also found that iron oxide strongly adsorbs Zn.
In summary, the components of the sediment are entangled with each other, shielding or overlapping adsorption sites, so the adsorption amount at each depth is not consistent with the OM content, CEC, iron and aluminium oxide contents or the change trend of the adsorption amount and the adsorption capacity does not increase with increases in OM content, CEC, or Fe-Al oxide content. The sediment components exhibit differential adsorption of heavy metals: iron-aluminium oxide contributes more to the adsorption of Cu( ), and OM and iron oxide contribute more to the adsorption of Zn( ).
Different components of sediment have different adsorption properties for heavy metals; iron and aluminium oxides contribute more than other components to Cu adsorption, while OM and iron oxide contribute more to Zn adsorption.
Previous research results showed that there is a certain linear positive correlation between the OM content, CEC, and metal oxide contents in sediments and the adsorption amount 18-20 . The results are shown in Table 8. The R 2 between each component and the amount of adsorption is not high, indicating that the amount of adsorption does not depend on a single component, but is the result of the combined effect of multiple components.
In the adsorption of Cu and Zn, as the equilibrium adsorption concentration increases, the competition for adsorbate molecules to occupy the adsorbent sites becomes more intense. When a high-binding-energy adsorption site is close to its full energy, non-speci c adsorption increases, and the adsorption rate gradually slows 21 . The iron-aluminium oxides and OM in group B, group C, and group D masked adsorption sites; thus, the K values were much lower in these groups than in group A (400.3). As the adsorption equilibrium concentration increased, the number of adsorption sites of groups B, C, and D decreased faster than those of group A, and the growth rate of the adsorption capacity slowed faster than that of group A. Relevant studies have shown 22 that as the equilibrium concentration increases, the adsorption capacity increases, and the adsorption curve shows a sharp increasing trend at low concentrations. However, the adsorption potential is limited, and when the concentration of heavy metal ions is high, the electrokinetic potential and electrical properties of colloidal particles are reduced, which reduces the stability of heavy metal-colloid-soil aggregates, and the curve gradually attens 23 . The decrease in adsorption capacity with the increase in adsorption equilibrium concentration may be caused by the decrease in adsorption sites and the limited adsorption capacity.
In practice, the surface of sediment particles is uneven, which makes the number and distribution of adsorption sites uneven. The adsorption isotherms of the sediments for Cu and Zn were more consistent with the Freundlich adsorption isotherm model than the Langmuir model, indicating that the adsorption process follows multi-layer adsorption. The surface of sediment particles is not uniform, so the t of the Freundlich isotherm adsorption model aligns with reality; Mustapha 's et al. 24 research results are similar. The order of the adsorption amounts of the samples corresponds to the order of their K values (because some samples could not be tted by the Langmuir isotherm adsorption model, the Freundlich adsorption isotherm model K value was used for comparison), which is similar to most research results 25  The organic/inorganic composite content was not equal to the sum of OM and metal oxides, presumably because hydrogen bonding, ion exchange and hydrophobic forces, such as anion adsorption mechanisms, embedded the composites in the mineral surface and between layers of swollen clay mineral crystals 27,28 , affecting the mineral cementation degree and thus colloid stability 29 . In addition to directly participating in the formation of complexes, the strong surface activity of iron and aluminium oxide can form bridges with OM and stabilize colloids through coordination exchange or the formation of ionic bonds 30 . In this way, these components combine to form organic and inorganic complexes that form the core and structure of the soil [31][32][33] .
To further compare the adsorption capacity differences between OM-iron-aluminium oxide composites, OM and iron-aluminium oxide, we compared the four groups of sediment samples under different pH values and adsorption quantity changes (Fig. 7). We found that at different pH values, the sediment Cu and Zn adsorption performance for group A was greater than that of the other three groups. Similar results were reported by Perez-Novo et al. 34 .
In group A, OM and iron and aluminium oxides formed organic-inorganic complexes, increasing the sediment surface area and surface activity and enhancing the adsorption capacity 35 . In addition, Mg(II) and Fe(II) compounds in sediments may be dissolved due to the presence of a large amount of H + in low-pH environments, thus competing for adsorption sites, or Cu could form hydroxylates with increasing pH 21,36,37 . In group D, OM and iron and aluminium oxide were removed simultaneously, which greatly reduced the number of active sites on the sediment surface and exposed the silicic skeleton of the sediments, making adsorption more susceptible to the in uence of pH value. In summary, the organic-inorganic composites in the sediments did not correspond to the simple addition of OM and iron-aluminium oxides. Their contribution rates to the adsorption of Cu and Zn were G OM−IAO = (G OM +G IAO ) *0.4-2 and G OM−IAO= (G OM +G IAO ) *1.18-3.35. When organic-inorganic complexes are formed, the zeta potential decreases, the surface negative charge increases, and the number of adsorption sites such as functional groups and variable charges increase, making the adsorption capacity of organicinorganic complexes for Cu and Zn signi cantly higher than those of OM and iron and aluminium oxides.

Conclusion
(1) Along the direction of water ow, with increasing distance from the river, the adsorption of Cu and Zn showed no signi cant change trend at a depth of 0-10 cm, a decreasing trend at 10-20 cm, and an increasing trend at 20-30 cm. In terms of depth, the adsorption capacity presented a trend of 10-20 cm > 0-10 cm > 20-30 cm.
(2) The components of the sediment were entangled with each other, shielding or overlapping adsorption sites, so that the adsorption capacity did not increase with increasing OM content, CEC, and iron aluminium oxide content. Different sediment components exhibit different adsorption properties of heavy metals. Iron and aluminium oxide contribute more than other components to the adsorption of Cu by deposition, while OM and iron oxide contribute more to the adsorption of Zn.

Competing interests
The authors declare no competing interests.
Funding Figure 1 Distribution of sampling points. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.