Oyster shell-modified lignite composite in globular shape as a low-cost adsorbent for the removal of Pb²⁺ and Cd²⁺ from AMD: Evaluation of adsorption properties and exploration of potential mechanisms

3.1.1 Preparation conditions of OSL

In Fig. 1(a), it is evident that the pyrolysis temperature of oyster shells played a crucial role in the removal of Pb²⁺ and Cd²⁺ by OSL. The removal rate and adsorption capacity of both pollutants increased significantly within the temperature range of 750 °C to 900 °C. At 900 °C, the removal rate of Pb²⁺ and Cd²⁺ reached 98.92 % and 99.08 %, respectively, and the adsorption capacity was 247.3 mg/g, and 49.54 mg/g, respectively. Although the removal rate and adsorption capacity slightly increased within the temperature range of 900 °C to 1000 °C, the increase was not significant. The presence of organic matrix in oyster shells may affect its pyrolysis temperature, but the content of organic matrix in oyster shells is relatively low, less than 5 %, so it is negligible (Xu et al., 2021). Oyster shell powder has the characteristics of a large surface area, small particle size, and rich porous structure, which reduces the thermal stability of calcite structure. The reasons for this phenomenon are as follows: Firstly, oyster shells, which are rich in CaCO₃, decompose into CaO after high-temperature pyrolysis. This process is accompanied by a large amount of CO₂ gas escaping, resulting in the transformation of the layered structure of oyster shells into a new porous structure with a large specific surface area. This increases the contact area between the main component CaO and water and heavy metals (Xu et al., 2019). Secondly, CaO has a strong electrostatic affinity for hydroxyl functional groups at its coordination position, making it easy to hydrolyze and complexly react with heavy metal ions. As a result, heavy metals can be removed (Shi et al., 2022). Thirdly, after ultrasonic treatment, a portion of CaO is hydrolyzed and converted to Ca(OH)₂. Ca(OH)₂ plays a role in activating and expanding the pores of lignite during the agitated impregnation process, introducing more hydroxyl and other functional groups on the surface of lignite, and improving the pore structure of lignite. This greatly enhances the ability of lignite to adsorb heavy metal ions (Moon et al., 2013; Zhong et al., 2021). Ha et al. (2019) found that the CaCO₃ and CaO contents of Taean and Tongyeong oyster shells were different when the pyrolysis temperature was between 700 °C and 1000 °C. Due to the difference in organic matter content, fineness, and calcite texture, the pyrolysis temperature of oyster shells in different regions is also different. Combined with the results of this experiment and previous studies, it was speculated that the pyrolysis of oyster shells used in this experiment at 900 °C achieved a large amount of decomposition of the calcite structure, lost CO₂, and transformed into CaO, which was also confirmed by Suwannasingha et al. (2022). The pyrolysis temperature of 900 °C was chosen for OSL preparation.

Close

Fig. 1

The preparation conditions (a) pyrolysis temperature, (b) pyrolysis time, (c) L/OS mass ratio of OSL and the preparation conditions (d and e) adhesive content, (f and g) roasting temperature, (h and i) roasting time of OSL-G.

Export to PPT

In Fig. 1(b), the removal rates and adsorption capacity of Pb²⁺ and Cd²⁺ by OSL increase as the pyrolysis time of oyster shells goes from 10 min to 20 min. However, there is no significant change in the removal rates and adsorption capacity when the pyrolysis time is further increased after 20 min. This suggests that the decomposition of oyster shells starts at 10 min and is mostly finished by 20 min. During this process, a majority of calcite is transformed into CaO. These findings are consistent with the results obtained by Xia et al. (2021). According to Alidoust et al. (2015), the diameter of oyster shells is not expected to decrease regardless of the pyrolysis temperature. However, an increase in pyrolysis time leads to the recrystallization of particles, resulting in larger particle sizes and reduced surface area. This, in turn, hinders the hydration of CaO to produce Ca(OH)₂, thereby reducing the effectiveness of OSL in removing Pb²⁺ and Cd²⁺. To prevent crystal aggregation after pyrolysis, ultrasonic shear treatment is recommended. To minimize energy consumption and processing costs during subsequent ultrasonic treatment, while ensuring the adsorption performance of OSL for Pb²⁺ and Cd²⁺ in AMD. The pyrolysis time of 20 min was chosen for OSL preparation.

In Fig. 1(c), when the mass ratio of lignite to oyster shells is 1:1, the removal rates and adsorption capacity of Pb²⁺ and Cd²⁺ by OSL are the highest, achieving 99.36 % and 100 %, 248.4 mg/g and 50 mg/g, respectively. As the proportion of oyster shells increased, the removal effect decreased. On one hand, the alkalinity and solubility of oyster shells are enhanced after undergoing pyrolysis and ultrasonic treatment. However, excessive loading of lignite can lead to an excessive amount of –OH on the surface of OSL, which accelerates the dissolution rate. Consequently, when metal ions in an aqueous solution come into contact with the OSL surface, they quickly precipitate and generate a significant amount of hydroxide. This, in turn, leads to substantial precipitation and accumulation on the OSL surface, ultimately blocking the pores of composite materials and hindering the adsorption reaction. On the other hand, oyster shells hold a significant quantity of Ca²⁺. The inclusion of suitable Ca²⁺ in OSL facilitates the ion exchange process with Pb²⁺ and Cd²⁺ in solution, effectively eliminating them (Khan et al., 2018; Madrid and Lanzon, 2017). Nevertheless, an excessive cation content on OSL generates repulsion towards heavy metal ions in solution, hindering their proximity to OSL (Li et al., 2015; Lu et al., 2018). Consequently, the adsorption capability of OSL on Pb²⁺ and Cd²⁺ is influenced. The mass ratio of lignite to oyster shells was 1:1 to prepare OSL.

3.1.2 Preparation conditions of OSL-G

The excellent properties of adsorbed materials not only depend on the adsorption effect but also consider their recyclability and secondary pollution problems. With this in mind, a spherical composite material, OSL-G, was prepared by using lignite and oyster shell as the substrate and adding bentonite as the binder. The composite material can release alkalinity slowly, and solve the problem of large dispersion and difficult separation and recovery of powdering adsorption material. Bentonite is a kind of natural clay mineral with good adhesion, plasticity, non-toxicity, and thermal stability. After roasting with lignite powder and oyster shell powder, a volcanic ash reaction will occur, resulting in the formation of stable and durable calcium silicate hydrate (C-S-H) or calcium aluminosilicate hydrate (C-A-S-H) (Zhan et al., 2020; Bai et al., 2021). This facilitates the slow release of alkalinity and the effective removal of Pb²⁺ and Cd²⁺ in AMD. Sun et al. (2020) found that the generation of hydration products decelerates the release rate of alkalinity, but too much gel would reduce the permeability of the material after hardening. The adhesive content plays a crucial role in determining the removal rate of Pb²⁺ and Cd²⁺, the loss rate of OSL-G, and the pH in AMD. In Fig. 1(d and e), it is evident that the removal rate and adsorption capacity of Pb²⁺ and Cd²⁺, the loss rate of OSL-G, and the pH of the solution system all decreased as the adhesive content increased. When the adhesive content was 15 %, the loss rate of OSL-G decreased to 9.74 %, while the removal rates and adsorption capacity of Pb²⁺ and Cd²⁺ remained at 99.87 %, and 96.71 %, 12.48 mg/g, and 2.42 mg/g, respectively. In addition, the pH of the solution system increased from 4 to 6.93 after the reaction. Under this condition, OSL-G can not only maintain a high removal rate of Pb²⁺ and Cd²⁺, but also ensure the strength of particles, and the pH of the reaction solution system can reach the emission standard. OSL-G was prepared with a selected adhesive content of 15 %.

After extruding the lignite and oyster shell powder with 15 % adhesive, roasting them at a specific temperature can enhance the strength of the particles and improve the balance force between them. Additionally, this process can also enhance the ability of the OSL-G composite to release alkalinity. Fig. 1(f and g) shows that the loss rate of OSL-G decreases significantly with the increase of roasting temperature in the range of 400 °C ∼ 600 °C. However, the loss rate increases with the increase of roasting temperature after 600 °C. According to the research results of Bai et al. (2021), when the roasting temperature rises from 400 °C to 500 °C, some SiO₂ in the material preferably transforms into an amorphous state and reacts to form C-S-H and other hydrates. Hydrate makes OSL-G have a certain mechanical strength, but its loss rate is still high due to insufficient production. The roasting temperature gradually increased from 500 °C to 600 °C, and SiO₂ and other Al and Si compounds in a large number of materials reacted with calcium oxides in oyster shells to form gel products C-S-H, C-A-S-H, etc., which was conducive to enhancing the strength of OSL-G (Okano et al., 2013; Alastair et al., 2018). As the temperature continued to rise, the gel product transformed into a crystalline phase, and the higher the temperature, the higher the crystallinity. However, the crystallization of the gel products C-S-H and C-A-S-H will cause the binding force between the materials to decrease, resulting in the loss rate increasing again. In addition, roasting will lead to material moisture loss and will lead to brown coal and oyster shell powder decomposition to produce CO₂ gas. The loss of water vapor and the escape of CO₂ gas will cause the porosity of OSL-G. The higher the roasting temperature, the faster the water vapor dissipation, and the stronger the decomposition reaction, resulting in excessive pores of OSL-G, the collapse of pore structure, and the decrease of mechanical strength. It can also be seen from Fig. 1(f and g) that with the increase in roasting temperature, the removal rate of Pb²⁺ by OSL-G fluctuates slightly but maintains a high state, while the removal rate of Cd²⁺ and the pH of the solution system obviously increase first and then tends to equilibrium. Roasting causes the decomposition of CaCO₃ in oyster shell powder, resulting in the formation of CaO. This increases the alkalinity of OSL-G. The addition of OSL-G to a solution helps raise its pH, facilitating the precipitation of Pb²⁺ and Cd²⁺ and ultimately achieving a higher removal rate. The likelihood of the CaCO₃ decomposition reaction occurring increases with higher calcination temperatures. However, as the temperature increases, the CaO and its hydrolyzed products react with the silicate and silica aluminate precursors in the material, leading to the formation of gelling substances. Excessive gelling products can then block the pore structure of the adsorbent material (Mo et al., 2018), impacting the removal of Pb²⁺ and Cd²⁺. OSL-G was prepared at a temperature of 600 °C.

In Fig. 1(h and i) it is evident that with the increase of roasting time, the loss rate of OSL-G decreases. This is because the extension of roasting time is not only conducive to the decomposition of CaCO₃ of oyster shells into CaO but also conducive to the full reaction among Ca, Al, and Si oxides to generate cementation products, thus enhancing the mechanical strength of OSL-G. It can also be seen from Fig. 1(h and i) that the removal rates and adsorption capacity of Pb²⁺ and Cd²⁺ by OSL-G increased first and then decreased with the increase of roasting time, and the pH of the solution system also has the same trend. During the roasting process, the production of silicate and silica aluminate compounds is insufficient to meet cementation requirements within the 0.5 h to 1.5 h timeframe. However, extending the roasting time in this range has two positive effects. Firstly, it promotes the formation of more alkaline CaO. Secondly, it facilitates the evaporation of water molecules and the volatilization of CO₂ gas, which is a decomposition product of organic matter and carbonate. This leads to an increase in the pore size of the adsorbed material. Consequently, a short-term increase in roasting time enhances the capacity for alkalinity release, resulting in an improved removal rate of Pb²⁺ and Cd²⁺. When the roasting time reaches 2 h, further increasing it will result in excessive cementation products of OSL-G and increased crystallinity. This, in turn, will make it difficult to release alkalinity, ultimately reducing the removal rate of Pb²⁺ and Cd²⁺. Comparatively, the solubility product constant of Cd²⁺ precipitate is smaller than that of Pb²⁺, making the removal of Cd²⁺ more influenced by the alkalinity of the solution system. Consequently, the adsorption effect of OSL-G on Cd²⁺ undergoes a more significant change with variations in roasting temperature. OSL-G was prepared with a roasting time of 2 h.

3.1.3 Optimize preparation conditions of OSL-G by RSM

The BBD levels, test design, and responses are shown in Table B1. Response surface regression analysis was performed on the experimental results using Design Expert V8.0.6 software. Quadratic regression equations were established between the adhesive content (A), roasting temperature (B), roasting time (C) of the independent variable and the removal rate of Pb²⁺ (Y₁), the removal rate of Cd²⁺ (Y₂), the loss rate of OSL-G (Y₃), and pH value of the solution (Y₄), as shown in Eqs. (1)–(4).

Analysis of variance (ANOVA) was performed on the above quadratic regression equation to determine the accuracy of the model and the significance of the influence of various factors and their interactions on the response value. The results are shown in Tables B2–B5. The significance of the model is determined by both the F-value and P-value. The larger the F-value is and the smaller the P-value is, the more significant the impact will be. The F values of the regression models from Y₁ to Y₄ are 26.29, 109.88, 243.27, and 181.06, respectively. The P values of these models are all < 0.0001, and their lack of fit is 0.8347, 0.6903, 0.1551, and 0.105, respectively. It shows that the model has high reliability, the model reaches a significant level, has statistical significance, and can be used to analyze the results of the actual point of the test (Mousavi et al., 2018; Luo et al., 2021). In addition, if the P-value of primary terms A, B, and C of the model is less than 0.05, it indicates that the three factors A, B, and C have significant influence on the response value, and vice versa. If the P-values of the secondary interaction terms AB, AC, and BC of the model meet the significance requirement, it indicates that the interaction effect between the secondary interaction terms is significant (Zhang et al., 2023a,b). The R² of each model is 0.9713, 0.9930, 0.9968, 0.9957, which is very close to 1, indicating that the regression model has a high degree of fit. In addition, the accuracy of the experiment was tested by three coefficients, namely, R_Adj² − R_Pred² < 0.2, coefficient of variation (C.V.) < 10 %, and Adeq Precision (AP) > 4. If all the requirements were met, the accuracy, reliability, and precision of the experiment could be considered to be high (Kacakgil and Cetintas, 2021; Saeed et al., 2021). Hence, the aforementioned regression models effectively describe the response values Y₁ ∼ Y₄, making them suitable for analysis, prediction, and optimization purposes. In the Pb²⁺ removal rate model, F values of A, B, and C were 100.35, 17.12, and 5.74, and P values were <0.0001, 0.0044, and 0.0478, respectively. The results showed that the adhesive content had a significant effect on the removal rate of Pb²⁺. The significant degree was the adhesive content > roasting temperature > roasting time. Similarly, in the other three models, the influence of adhesive content and roasting temperature on the removal rate of Cd²⁺, the loss rate of OSL-G, and the pH of the solution was found to be highly significant. The significant degree was the adhesive content > roasting temperature > roasting time.

To comprehensively consider the influence of three factors A, B, and C on the response value, a graphical analysis was carried out, and the three-dimensional response surface map was shown in Fig. 2. It can be seen from Fig. 2(a), in the Pb²⁺ removal rate model, the contours of the interaction between AB and AC present a long oval shape with long axis, indicating that the interaction effect between AB and AC is significant. The distribution density of the contours on the X-axis is significantly higher than that on the Y-axis, indicating that the main effect of A is larger than that of B and C. However, the interaction between B and C is not significant, showing that the contour map presents a circular distribution, and the three-dimensional response surface map is steeper. Similarly, it can be seen from Fig. 2(b) that the same results are also obtained in the Cd²⁺ removal rate model. It can be seen from Fig. 2(c) that in the loss rate model, the interaction of AB is not significant, while the interaction of AC and BC is significant. It can be seen from Fig. 2(d) that in the pH value model, the interaction between AB and BC is significant, while the interaction between AC is not. The results of ANOVA also fully verified the above conclusions.

Close

Fig. 2

The three-dimensional response surface maps of (a) removal rate of Pb²⁺, (b) removal rate of Cd²⁺, (c) loss rate of OSL-G, (d) pH of the solution.

Export to PPT

To prepare OSL-G composite with good adsorption effect of Pb²⁺ and Cd²⁺ and low loss rate, the Design expert V8.0.6 software was used to optimize the preparation method of OSL-G composite. The results showed that the optimum preparation conditions of OSL-G composites were as follows: the adhesive content was 15.48 %, the roasting temperature was 642 °C, and the roasting time was 2.37 h. It is predicted that the removal rate of Pb²⁺ and Cd²⁺ were 99.89 % and 99.02 %, the loss rate was 9.87 %, and the pH of the solution was 7.49. Considering the feasibility and economic cost of the experiment, some adjustments were made to the experimental results. The optimum preparation conditions of OSL-G were determined the adhesive content was 15.50 %, the roasting temperature was 640 °C, and the roasting time was 2.35 h.

3.3.1 Effect of adsorbent dosage

In Fig. 4(a), it is evident that when the adsorbent dosage is between 1 and 2 g/L, the OSL-G adsorption site can be fully utilized, resulting in an increase in the removal rate of Pb²⁺ from 48.48 % to 98.12 %. Further increasing the dosage has minimal impact on the removal rate of Pb²⁺. Thus, the optimal dosage for OSL-G adsorbing Pb²⁺ is considered to be 2 g/L. For the adsorption of Cd²⁺, increasing the dosage of OSL-G from 1 g/L to 4 g/L leads to an increase in the removal rate of Cd²⁺ from 30.21 % to 99.45 %. Continuing to increase the dosage has little significance, indicating that the optimal dosage for OSL-G adsorbing Cd²⁺ is considered to be 4 g/L. The improvement in adsorption can be attributed to the increase in surface area and the availability of more adsorption-binding sites (Pérez et al., 2022; Zhang et al., 2019). The difference in dosage required for the effective removal of Pb²⁺ and Cd²⁺ by OSL-G is related to their difference in radius, hydration energy, and solubility product constant. Compared with Cd²⁺, Pb²⁺ has a smaller radius, lower hydration energy, smaller solubility product constant, and is more prone to precipitation, ion exchange, and complexation reactions. Therefore, the optimal dosage for the adsorption of Pb²⁺ by OSL-G is only 2 g/L, while the optimal dosage for the adsorption of Cd²⁺ is up to 4 g/L.

Close

Fig. 4

Effect of the adsorbent dosage (a), initial pH value (b), initial concentration (c), binary metal system (d) on the adsorption of Pb²⁺ and Cd²⁺ by OSL-G.

Export to PPT

3.3.2 Effect of initial pH

It can be seen from Fig. 4(b), with the initial pH value increasing from 2 to 4, the removal rate of Pb²⁺ by OSL-G significantly increased from 38 % to 98.39 %, and the removal rate of Cd²⁺ increased from 18.06 % to 99.45 %. A similar behavior was found by Du et al. (2011) using two types of mollusk shell powders to remove Pb²⁺, Cd²⁺ and Zn²⁺. As the initial pH continued to increase from 4 to 6, the removal rate of Pb²⁺ and Cd²⁺ by OSL-G increased slightly and tended to be stable. The initial pH of the solution system will affect the electrostatic interaction, surface precipitation, and adsorption reaction between the two (Liu et al., 2009). The increase of initial pH and the low concentration of H⁺ is conducive to reducing the competition between H⁺, Pb²⁺, and Cd²⁺, leaving more adsorption sites remaining on the surface of OSL-G and removal of target pollutants. The high concentration of OH⁻ in the solution will increase the negative charge on the surface of OSL-G, thereby reducing the electrostatic repulsion of OSL-G with Pb²⁺ and Cd²⁺. It is also conducive to the neutralization and precipitation or the existence form of multiple basic ions.

It can be seen from Fig. 4(b) that the initial pH value is in the range of 2 ∼ 5, and the pH of the solution system after the reaction is all within 9. The pH range of AMD that OSL-G can handle is more extensive. In addition, pH affects the reaction mechanism of removing Pb²⁺ and Cd²⁺, and the stability of compounds generated by different reaction mechanisms is different (Pehlivan et al., 2008; Li et al., 2017). When pH < 7, Pb²⁺ and Cd²⁺ exist as ions, and the removal of metal ions by the composite is mainly based on ion exchange and surface complexation. When the pH of the solution is between 7 and 9, Pb²⁺ and Cd²⁺ mainly exist in the form of Pb(OH)⁺, Cd(OH)⁺, Pb(OH)_2, and Cd(OH)₂, and the metal removal of the composite is mainly precipitated. When the solution pH > 9, Pb²⁺ and Cd²⁺ mainly exist in the form of Pb(OH)₂, Pb(OH)₃⁻, Pb(OH)₄²⁻, Cd(OH)₂, Cd(OH)₃⁻ (Kolodynska et al., 2012; Jiang et al., 2020). The removal of metals by composite materials is mainly caused by precipitation, but the high pH value will lead to the secondary dissolution of hydroxide precipitation (Frišták et al., 2015). Therefore, considering the adsorption effect and the product stability of the adsorption reaction, it is considered that the initial pH value of 4 is more advantageous.

3.3.3 Effect of initial concentrations of Pb²⁺ and Cd²⁺

It can be seen from Fig. 4(c), that in the same temperature system, as the initial concentrations of Pb²⁺ and Cd²⁺ increase, the removal rates of Pb²⁺ and Cd²⁺ by OSL-G gradually decrease. This is because at a low initial concentration, compared with the amount of Pb²⁺ and Cd²⁺ in the solution, the number of effective adsorption sites on the surface of OSL-G is quite sufficient, which is conducive to the diffusion of Pb²⁺ and Cd²⁺ to the adsorption site and binding with it to promote the adsorption reaction (Jamshidifard et al., 2019). The concentration difference between the solution and the surface of the composite material increases, which strongly drives the Pb²⁺ and Cd²⁺ in the solution to move rapidly to the OSL-G surface, increasing the chance of interaction between Pb²⁺ and Cd²⁺ and the active site (Awang et al., 2019; Liu et al., 2020). However, with the advancement of the reaction process, the fewer the remaining adsorption sites, the adsorption gradually reaches saturation, and after the adsorption of Pb²⁺ and Cd²⁺, the surface of the composite material will accumulate excessive positive charges, which is difficult to continue adsorption with more free metal ions in the solution, so the removal rate is gradually reduced. At the same concentration level, the removal rates of Pb²⁺ and Cd²⁺ by OSL-G increased with the rise in ambient temperature, suggesting that the adsorption process of Pb²⁺ and Cd²⁺ by OSL-G is endothermic. As the temperature increases, the proportion of activated ions in the solution also increases. This leads to a faster diffusion rate of Pb²⁺ and Cd²⁺ towards the surface of the composite material. Consequently, there is a higher likelihood of effective collision with the functional groups on the surface of the composite material, resulting in an increased level of adsorption equilibrium. Similar results have also been described in previous articles (Ogata et al., 2018; Arslanoglu et al., 2019; Cherdchoo et al., 2019).

According to the “Comprehensive Sewage Discharge Standard” (GB 8978–1996), the maximum allowable emission limits for Pb²⁺ and Cd²⁺ are 1 mg/L and 0.1 mg/L, respectively. In this experiment, the initial concentrations of Pb²⁺ and Cd²⁺ in simulated AMD were 50 mg/L and 10 mg/L, respectively. To meet the AMD emission standard after treatment, the removal rates of Pb²⁺ and Cd²⁺ should reach 98 % and 99 %, respectively. Comparing the removal rates of Pb²⁺ and Cd²⁺ by OSL-G at different initial concentrations and temperatures, it was found that the removal rates of Pb²⁺ can exceed 99 % when the initial concentration is between 10 and 100 mg/L. However, when the initial concentration of Cd²⁺ increased from 10 mg/L to 30 mg/L, the removal rate dropped significantly from more than 99 % to less than 70 %. In summary, 2 g/L OPL-G can treat AMD with an initial concentration of Pb²⁺ ranging from 10 to 100 mg/L, and 4 g/L OPL-G can treat AMD with an initial concentration of Cd²⁺ ranging from 0 to 10 mg/L. Since the content of Cd in acid mine wastewater is generally low and the content of Pb is high, the treatment of Pb and Cd pollution in AMD by OSL-G has certain advantages.

3.3.4 Effect of the binary metal system

In Fig. 4(d), it is evident that the adsorption effect of OSL-G on Pb²⁺ and Cd²⁺ in the binary metal system is lower than that in the monadic metal system, and the higher the initial concentration is, the more significant the competitive effect of Pb²⁺ and Cd²⁺ is. This is because in the binary system, the total amount of metal ions in the solution is higher than that in the monadic metal system, and the points that can be provided by adding the same amount of materials are limited, resulting in competition between ions (Huang et al., 2020; Zaman et al., 2023). On the whole, the removal rate of Pb²⁺ is higher than that of Cd²⁺, indicating that OSL-G has a higher selectivity and better removal effect on Pb²⁺ in a non-equal competition system. The selective adsorption of heavy metal ions depends on the physical and chemical properties of metal ions (hydrolysis constant, radius of hydrated ion, molar mass, and solubility product constant of metal ion hydroxide). The smaller the hydrolysis constant of metal ions is, the easier the ions are to be adsorbed. The hydrolysis constant of Pb²⁺ (7.8) is significantly lower than that of Cd²⁺ (9.2). For homogeneous metal cations, the smaller the hydration radius, the easier it is to diffuse through the boundary layer to the surface of the adsorbed material. The hydration radius of Pb²⁺ (0.401 mm) is smaller than that of Cd²⁺ (0.426 mm), the mass transfer resistance of Pb²⁺ in solution is smaller, and the ion exchange reaction is more likely to occur. The higher the molar mass of metal ions, the more likely a precipitation reaction occurs. The molar mass of Pb²⁺ (207.2 g/mol) is almost twice that of Cd²⁺ (112.4 g/mol), and the adsorption preference of Pb²⁺ is stronger. The lower the solubility of hydroxide of the same valent metal ion, the easier it is to react with OH⁻ to form chemical precipitation. The hydroxide solubility product constant of Pb²⁺ (1.2 × 10⁻¹⁵) is smaller than that of Cd²⁺ (2.2 × 10⁻¹⁴). Therefore, it can be concluded that the selection order of Pb²⁺ and Cd²⁺ by OSL-G is Pb²⁺ > Cd²⁺, which is also verified by similar research results (Wan et al., 2014; Bohli et al., 2015; Gameli et al., 2023).

It can also be seen from Fig. 4(d) that in the Pb-Cd binary metal system (Cd²⁺=10 mg/L, Pb²⁺ = 10–100 mg /L), the removal effect of OSL-G on Pb²⁺ remains almost unchanged, both remaining above 99 %. In the Cd-Pb binary metal system (Pb²⁺ = 10 mg/L, Cd²⁺ = 10 mg/L), the effect of Pb²⁺ competitive adsorption on the removal of Cd²⁺ by OSL-G adsorption is also low, and the removal rate of Cd²⁺ can be ensured above 99 %. Therefore, adding 4 g/L OSL-G to treat the simulated AMD-containing binary metal system (pH = 4, Pb²⁺ = 10 mg/L, Cd²⁺ = 10 mg/L) can effectively remove 99.38 % of Pb²⁺ and 99.02 % of Cd²⁺, and the pH of the solution system after reaction is around 7. The effluent water quality conforms to the “Comprehensive Sewage Discharge Standard” (GB 8978-1996).

3.4.1 Adsorption kinetics

The adsorption process can generally be described as three processes: the diffusion on the surface of the adsorbent, the diffusion in the adsorbent particles, and the adsorption on the porous adsorption site. The time consumed at each stage affects the rate of the entire adsorption process (Li et al., 2019). Quasi-first-order kinetic and quasi-second-order kinetic adsorption models (Fig. 5(a and b)) and inter-particle diffusion models (Fig. 5(c and d)) were used to fit the adsorption processes of Pb²⁺ and Cd²⁺ by OSL-G, determine the adsorption rate control steps and explore the adsorption mechanism. The adsorption parameters of the kinetic models are shown in Table 3.

Close

Fig. 5

Adsorption kinetics, isotherms, thermodynamics of Pb²⁺ and Cd²⁺ by OSL-G. (a) Adsorption kinetics of Pb²⁺, (b) Adsorption kinetics of Cd²⁺, (c) Intra-particle diffusion model of Pb²⁺, (d) Intra-particle diffusion model of Cd²⁺, (e) Langmuir model of Pb²⁺, (f) Langmuir model of Cd²⁺, (g) Freundlich model of Pb²⁺, (h) Freundlich model of Cd²⁺, (i) Effect of temperature on the adsorption of Pb²⁺ and Cd²⁺ by OSL-G, (j) Adsorption thermodynamics of Pb²⁺, (k) Adsorption thermodynamics of Cd²⁺.

Export to PPT

It can be seen from Fig. 5(a and b) that, compared with the quasi-first-order kinetic model, the R² of the quasi-second-order kinetic model is closer to 1, and the theoretically calculated value of the adsorption capacity of Pb²⁺ and Cd²⁺ by OSL-G obtained by this model is closer to the actual experimental value. The parameter results in Table 3 show that the quasi-second-order kinetic model can more accurately reflect the adsorption process. The establishment theory of the quasi-second-order kinetic model assumes that the adsorption rate is controlled by the chemisorption mechanism and is determined by the square of the number of remaining adsorption vacancies on the surface. Therefore, it can be considered that the adsorption rate is directly proportional to the quadratic of the driving force. In addition, during the adsorption process, Pb²⁺ and Cd²⁺ will react with the internal substances of OSL-G particles to form new chemical bonds, which will transform them into new substances and remove them, and their adsorption rate is mainly controlled by chemisorption. It can be seen from Fig. 5(c and d), there are three stages in the adsorption process of Pb²⁺ and Cd²⁺ by OSL-G. The first stage is the rapid diffusion of Pb²⁺ and Cd²⁺ through the boundary film to the surface of OSL-G, which is reflected as surface diffusion. The second stage is the slow diffusion of Pb²⁺ and Cd²⁺ to the mesopore in OSL-G, and the third stage is mainly the micropore diffusion, and the adsorption process is slow until the adsorption equilibrium is reached (Liang et al., 2018). The fitting parameters in Table 3 show that the order of diffusion rates of Pb²⁺ and Cd²⁺ is K₃₁ > K₃₂ > K₃₃, and the order of boundary layer thickness is C₁ < C₂ < C₃, which indicates that the boundary layer thickened continuously with the advancement of the adsorption process of Pb²⁺ and Cd²⁺ by OSL-G. The thickening of the boundary layer obstructs the diffusion, leading to the deviation of the adsorption process from the internal diffusion model. The fitted curves do not pass through the origin, and the deviation degree is increasing, which also verifies the above analysis results. The deviation of the inter-particle diffusion model from the origin indicates that the adsorption process is controlled by the external and internal diffusion of the particles (Wang et al., 2019). The low R² of the fitted inter-particle diffusion model also indicates that the entire adsorption process is mainly dominated by chemical adsorption.

3.4.2 Adsorption isotherm

The adsorption isotherm can effectively determine the distribution of adsorbents in the solid–liquid phase, which is of great significance for evaluating the adsorption process of Pb²⁺ and Cd²⁺ by OSL-G (El Atouani et al., 2019). The Langmuir adsorption isotherm model (Fig. 5(e and f)) and Freundlich adsorption isotherm model (Fig. 5(g and h)) were used to fit the adsorption processes of Pb²⁺ and Cd²⁺ by OSL-G. The fitting parameters of the isotherm model are shown in Table 4. As shown in Fig. 5(e–h) and Table 4, R2 of the Langmuir model is smaller than that of the Freundlich model at 288.15 K, 298.15 K, and 308.15 K temperatures. It can be seen that the adsorption of Pb²⁺ and Cd²⁺ by OSL-G at different temperatures is more suitable to be explained by the Freundlich adsorption isothermal model, indicating that the adsorption process is multi-layer non-uniform adsorption (Homagai et al., 2022). The K_F values of Pb²⁺ and Cd²⁺ adsorbed by OSL-G increased slightly with increasing temperature, indicating that the affinity of OSL-G to Pb²⁺ and Cd²⁺ was improved by increasing temperature. n reflects the strength of adsorption, and 1/n is the concentration index commonly used to evaluate adsorption. The smaller the value, the better. The 1/n values are all between 0 and 1, indicating that the adsorption reaction is easy to carry out (Kim and Singh, 2022).

The adsorption capacity of OSL-G and OSL for Pb²⁺ and Cd²⁺ in aqueous media was compared with that of other adsorbent materials (Table 5). It can be seen from Table 5, the maximum adsorption capacity of OSL-G for Pb²⁺ and Cd²⁺ was in the middle level, especially the adsorption capacity of Cd²⁺ was not high. Through further analysis of this phenomenon, it was found that the adsorption time, temperature, and adsorbent morphology were the key factors affected the adsorption capacity, especially the adsorbent morphology. Other materials with high adsorption capacity in Table 5 were powdery with small particle size, while OSL-G was granular, which also become an important reason for the low adsorption capacity of OSL-G. Therefore, based on the importance of adsorbent morphology, the adsorption capacity of powdered OSL was investigated, and it was found that the adsorption capacity was greatly improved, which was sufficient to prove the excellent performance of the adsorption material. Interestingly, although the powder adsorption effect is good, but compared with the powder adsorbent, the granular adsorbent has the advantage of easy solid–liquid separation and sedimentation. The effect of different morphologies of adsorbents on pollutants in water has also become an important direction of follow-up research.

3.4.3 Adsorption thermodynamics

The temperature has a certain effect on the adsorption reaction. It is necessary to explore the thermodynamic changes in the adsorption process of OSL-G to determine the spontaneity of the adsorption of Pb²⁺ and Cd²⁺. The adsorption experiments were carried out at three temperature conditions of 288.15 K, 298.15 K, and 308.15 K, respectively (Fig. 5(i)), and thermodynamic adsorption curves were drawn with lnK_d as Y-axis and 1/T as X-axis (Fig. 5(j and k)). The thermodynamic parameters of the adsorption process are shown in Table 6. It can be seen from Fig. 5(i–k), that the removal efficiency of Pb²⁺ and Cd²⁺ showed an increasing trend with the increase of temperature. In Table 6, the Gibbs free energy ΔG is negative at all temperatures, indicating that the adsorption of Pb²⁺ and Cd²⁺ by OSL-G is spontaneous (Sun et al., 2015). The higher the temperature, the smaller the ΔG, and the more likely the spontaneous reaction is to occur. The enthalpy change ΔH is positive, which also proves that the adsorption reaction is an endothermic process. The entropy change ΔS is positive, indicating that the adsorption process is a process of entropy increase, which means that the adsorption process of Pb²⁺ and Cd²⁺ is inevitably accompanied by the desorption of other ions at the adsorption site on the surface of OSL-G, which increases the chaos and disorder of the whole system during the adsorption process (Lin et al., 2018).

3.5.1 SEM

It can be seen from Fig. 6(a), that the surface of lignite is relatively smooth and dense, but there are also some micropores and lamellar accumulations on the surface, with some folds and voids (Zhang and Chen, 2020; Jellali et al., 2021). There are many cracks, pores, and scaly structures on the surface of OSL-G, and a large number of fine particles, flake structures, and massive structures accumulate. High-temperature roasting results in the release of water and some organic matter in lignite, the pore structure shrinks and collapses, and the surface roughness increases (He et al., 2019). The scale structure of the bentonite binder is retained, which makes the structure of OSL-G loose and porous. Oyster shells decompose into CaO and CO₂ when roasted. This analysis is also supported by the significant reduction of the content of C and O in OSL-G compared with that in lignite. A large amount of CO₂ escape will play a role in opening pores and increasing the number of pores in OSL-G. At the same time, the CaO generated by roasting at high temperatures will enter the bentonite and lignite layers and combine with SiO₂, silicate precursors or alumino-silicate precursors to form C-S-H, C-A-H, C-A-S-H and other crystals or amorphous gels, which are deposited in the form of fine particles, flake structure or block structure to increase the surface area of the material (Wang et al., 2011; Li et al., 2012; Hao et al., 2019). After adsorption of Pb²⁺ and Cd²⁺, OSL-G (Pb²⁺) and OSL-G (Cd²⁺) are attached to a large number of substances with different forms, and the aggregation state increases. Compared with the EDS diagram, it can be seen that there are no Pb and Cd elements in OSL-G, while Pb and Cd elements appear in OSL-G (Pb²⁺) and OSL-G (Cd²⁺), which indicates that the adsorption sites of OSL-G successfully interact with Pb²⁺ and Cd²⁺, and the products adhere stably to their surfaces.

Close

Fig. 6

Microscopic characterization results of Pb²⁺ and Cd²⁺ adsorbed by OSL-G. (a) SEM-EDS, (b) FTIR, (c) XRD.

Export to PPT

3.5.2 FTIR

It can be seen from Fig. 6(b), in the peak spectrum of lignite, the –OH stretching vibration absorption peak corresponding to 3402 cm⁻¹ and the –OH bending vibration absorption peak corresponding to 1630 cm⁻¹ both weaken after granulation, which can be attributed to the loss of water in lignite during roasting (Zhang et al., 2016). The absorption peaks at 1269, 2923, and 2850 cm⁻¹ were derived from the residue of organic matter in lignite material, and the peaks disappeared in the OSL-G peak spectrum after roasting. In the OSL peak spectrum, the in-plane and out-plane deformation vibrations of C-O and the stretching and bending vibrations of Ca-O appear near 708, 870, and 1408 cm⁻¹, indicating the presence of Ca-O and CO₃²⁻ in OSL-G (Feng et al., 2022). By comparing the peak spectra of lignite and OSL-G, it is found that there are Si-O-Si characteristic absorption peaks near 474 and 1020 cm⁻¹, and OSL-G is significantly enhanced compared with the Si-O-Si characteristic absorption peaks in lignite, which means that a new silicate or silicon oxide appears in OSL-G. By comparing the peak spectra before and after adsorption of Pb²⁺ and Cd²⁺, it is found that the absorption peaks near 474 and 1020 cm⁻¹ are enhanced in different amplitude, indicating that Si-O-Si participated in the adsorption reaction process. In addition, the absorption peaks of 708, 870, and 1408 cm⁻¹ still existed, indicating that the adsorption did not destroy the structure of OSL-G, but the intensity of the absorption peak increased to different degrees, indicating that Ca-O and C-O participated in the adsorption of Pb²⁺ and Cd²⁺.

3.5.3 XRD

It can be seen from Fig. 6(c), that lignite is mainly composed of organic carbon molecules and inorganic minerals. Lignite has an obvious characteristic peak at 2θ = 26.603°, which is the (0 0 3) plane diffraction peak of carbon. This proves that lignite is an amorphous carbon structure (He et al., 2019). The (1 0 0) plane diffraction peaks of SiO2 appear at 2θ = 20.859° and 26.639°, indicating that the minerals of lignite are composed of quartz (Abdelsayed et al., 2018). The EDS results showed that lignite still contained a certain amount of Al element, but no aluminum compounds were found in the phase detection because of its low content and small crystal structure. The characteristic peak of SiO₂ in OSL-G disappeared because part of SiO₂ participated in the reaction to produce CaCO₃, CaSi₂O₅, CaAl₂Si₂O₈, and Ca₇(Si₆O₁₈)(CO₃)·2H₂O mineral phases, and the other part of SiO₂ became an amorphous gel. CaCO₃ is the main component of calcite in oyster shells. During the sintering process at 640 °C, the middle part of CaCO₃ will decompose into CaO. Part of CaO will react with SiO₂, the main component of lignite, to form CaSi₂O₅, while the other part of CaO will undergo a hydrolysis reaction with surrounding water and be transformed into Ca(OH)₂. Ca(OH)₂ will continue to react with SiO₂, aluminum-containing oxides, and other volcanic ash to form CaAl₂Si₂O₈ and Ca₇(Si₆O₁₈)(CO₃)·2H₂O.

The main mineral phases of OSL-G (Pb²⁺) are CaCO₃, PbO, Pb₈Si₂O_12, and CaPb(OH)₄. Compared with OSL-G before adsorption, CaCO₃ still exists after the reaction, and a new mineral phase appears during the adsorption of Pb²⁺. The new product PbO after the reaction is caused by the precipitation of Pb(OH)₂ generated by the reaction of Pb²⁺ and OH⁻ during the collection and drying process. Pb₈Si₂O₁₂ is thought to be produced by ion exchange reaction between Pb²⁺ and silicates in OSL-G or electrostatic adsorption with free silicates in solution. CaPb(OH)₄ is the product of synergistic precipitation of Ca²⁺, Pb²⁺, and OH⁻. After OSL-G adsorbed Cd²⁺, CaCO₃ still existed, and new mineral phases including Cd(OH)₂, CdO, CdSO₄·H₂O, Cd₂SO₄(OH)₂, Cd₃SiO₅, Cd₂Al₂Si₂O₉ were generated. The Cd(OH)₂ generated after the reaction is the product of the precipitation reaction between Cd²⁺ and OH⁻, and the CdO is the product of the thermal decomposition of Cd(OH)₂, which indicates that OSL-G can release a certain amount of OH⁻. CdSO₄·H₂O is formed after the reaction, which is caused by an ion exchange reaction between Cd²⁺ and metal sulfate existing in lignite, and a complex reaction between Cd²⁺, OH⁻ and SO₄²⁻. The reasons for the generation of Cd₂SO₄(OH)₂ are relatively complex, partly due to the complexation of Cd²⁺, OH⁻ and SO₄²⁻, and partly due to the precipitation and accumulation of Cd(OH)₂ generated by the adsorption and precipitation of OSL-G with dissolved Cd²⁺ in solution, resulting in a large number of new pores. And then continue to adsorb Cd²⁺, SO₄²⁻, OH⁻ produced by the water. Cd₃SiO₅ is thought to be the result of ion exchange between Cd²⁺ and silicates in OSL-G or electrostatic adsorption with silicates in solution. The mineral phase of Cd₂Al₂Si₂O₉ is formed by the reaction of Cd²⁺, Al³⁺, and silicate ions. Similar results were obtained in the studies of Bartczak et al. (2018), Hao et al. (2019) and Wang et al. (2022).

In conclusion, there are four potential mechanisms for the adsorption of Pb²⁺ and Cd²⁺ by OSL-G (Fig. 7). i, electrostatic adsorption. ii, Pb²⁺ and Cd²⁺ generate metal oxide precipitates on their surfaces. iii, the ion exchange between Pb²⁺ and Cd²⁺ and their surface Ca²⁺. iiii, the surface complexation mechanism of Pb²⁺ and Cd²⁺ with their surface oxygen-containing functional groups.

Close

Fig. 7

The potential mechanism of the adsorption for Pb²⁺ and Cd²⁺ by OSL-G.

Export to PPT