Preparation of structured biochar, its adsorption capacity of N and P and its characterization

Water eutrophication is one of the most common ecological and environmental problems in the world (Farshad 2021). Nitrogen (N) and phosphorus (P) discharge into the natural water bodies in a large amount will lead to a serious threat to the ecological environment of the water (Chen et al. 2020; Dong et al. 2021). A high concentration of N and P will stimulate the rapid growth of algae in the water, leading to the outbreak of algal bloom and red tide, resulting in the excessive consumption of dissolved oxygen in the water and the death of fish and shrimps, and worsening the water quality (Ding et al. 2021; Wang et al. 2021; Yang et al. 2021). Extensive studies have been done on the technologies of N and P removal in the water bodies, mainly involving precipitation (Quan et al. 2010), film technology (van Voorthuizen et al. 2005), reverse osmosis (Karami et al. 2020), coagulation (Liao et al. 2013), flocculation (Sadri Moghaddam et al. 2010), catalytic reduction (Yang & Lee 2005), and adsorption (Perumal & Sankaran 2019). Among them, adsorption is a more efficient and selective technology in N and P removal. Biochar is a carbon-rich solid formed by pyrolysis at moderate-high temperature in a limited oxygen atmosphere (Mohammadi et al. 2021). It is widely used for the adsorption of N and P due to its rich pore structure and a large number of functional groups (Xiao-fei et al. 2016; Hongbo et al. 2017). However, powder biochar in water treatment is not easy to separate and difficult to recycle and reuse; in contrast, biochar with a certain size and strength can make some improvements (Joan et al. 2014; Hodgson et al. 2016; Qumber et al. 2018).

Heavy metals and pathogenic bacteria in the sludge may contaminate crops and damage human health when directly returned to the field, while harmful substances may ooze out and pollute groundwater and land when directly buried (Chang et al. 2020; Kou et al. 2021). Pyrolysis sludge can not only kill parasites, pathogens and microorganisms, but also realize carbon fixation. However, there is a large amount of organic material (about 50–70%) in the sludge, and biochar obtained from sludge is characterized by developed pores, but relatively few active sites on the surface (Zeng et al. 2018; Difang et al. 2019). Cattail (Typha angustifolia) is an aquatic macrophyte, widely used in wetland ecological restoration, and it is characterized by leafy straws, a big spongy intercellular space (Kuno et al. 2018; Hou et al. 2020). Biochar prepared from sludge and cattail (CS) via pyrolysis can make some improvements. The material used to prepare biochar is one of the important factors that determine its performance (Hong et al. 2017). For example, straw and sawdust are mainly composed of cellulose, hemicellulose and lignin, with high carbon content and low ash content, and the prepared biochar has developed pores. The specific surface area is large and the adsorption capacity is strong. Co-pyrolysis of straw or wood chips with sludge may promote the thermal conversion of sludge and improve the performance of sludge biochar.

The study found that the larger the molecular weight of the binder used to prepare the structural biochar, the less likely it is to block the micropores (Fang et al. 2015). The use of high-density polyethylene (HDPE) as the binder for the preparation of structural biochar has the characteristics of simple process and good performance. The research on the modification of eutrophic water body for this type of structural biochar is still lacking. In this study, the optimal parameters for the preparation of structural biochar from CS biochar and HDPE, and the adsorption properties and mechanisms were analyzed. The specific objectives were: (i) the optimal parameters for the preparation of structured biochar (SC), (ii) the adsorption properties and mechanism of structured biochar.

Preparation of CS: large macrophyte cattails and sludge were used as raw materials for the preparation of biochar. The cattails and sludge in the ditch of Nanyuan section of Jinzhi river in the upper reach of Dianchi Lake were collected in August 2020. A slow pyrolysis method was used. A certain volume of 300 g L⁻¹ KOH with an immersion ratio of 4 ml g⁻¹ was mixed the mixture of cattails and sludge (6:4 weight ratio) in a THZ-Q table freeze thermostat oscillator (THZ-Q Desktop Constant Temperature Refrigerating Oscillator, Jiangsu, China). After 6 h of oscillation, the mixture was put in a muffle furnace (ELF 11/14B Carbolite Ltd, UK) with the temperature rising to 500 °C at a rate of 10 °C min⁻¹, and carbonized for 0.5 h at 500 °C. The product was crushed and ground, then put into a sealed bag through a 60-mesh sieve and stored in a dryer for later use.

Preparation of SC: 2 × 2 × 2 cm cube-shaped CS was prepared (initial parameters: molding pressure 11 N (cm²)⁻¹; molding temperature 160 °C; molding time 95 min). A single factor experiment (as shown in Table 1, the experiment was repeated 3 times) was used to determine the optimal parameters for SC preparation. Thus, the preparation method was determined as follows: a mixture of HDPE (200 mesh) and CS at a mass ratio of 3:2 was preformed in the mold at a pressure of 5 N (cm²)⁻¹, heated at 160 °C for 95 min, then cooled and finalized, and stored in a dryer for later use.

Use 0.1 mol L⁻¹ NaOH and HCl to adjust the pH of 100 ml NH₄Cl and KH₂PO₄ solution with a concentration of 20 mg L⁻¹ to 2–12. The measurement of pHpzc: Replace the solution with the same volume of intial pH as 2–12 of deionized water, shake at 25 °C for 48 h, and measure the pH of the deionized water again.

The element content (EA) is measured by Thermo Flash2000 instrument. The contact Angle is measured by AST VCA Optima XE. Atomic force microscope (AFM) Brook Dimension Edge is adopted for surface topography. The surface crystal is measured by X-ray diffractometer (XRD) Brook D8. Surface functional groups is measured by X-ray photoelectron spectroscopy (XPS) Thermo Fisher Scientific K-Alpha and Thermo Scientific Nicoler IS5 Fourier Infrared Spectrometer (FTIR).

The characteristics of influence of SC with different sizing amounts on the adsorption capacity of P and N were similar (Figure 1), trending a decrease at first and an increase afterward. The increase of polyethylene content inhibited the adsorption of N on SC, especially when the polyethylene content was 40 and 50%, the adsorption amount of P and N was significantly different, and the adsorption amount of P on SC was higher than that of N. With the increase of sizing amounts, the mechanical strength of SC was enhanced (Figure 2). There was little difference in mechanical strength of SC between different adsorbates, indicating a similar effect of P and N solutions on the structure of SC. Different molding pressures have different effects on the adsorption capacity of P and N (Figure 1). With the increase in molding pressure, the adsorption capacity of P on SC decreased, while the adsorption capacity of N increased first followed by a decrease. It suggested the modification role of polyethylene in CS, which had changed some of physical properties of CS and improved the adsorption of N on CS as a result. This point was further explored in the analysis of characteristics of SC and CS. With the increase in molding temperature, the adsorption capacity of P and N on SC increased first and then decreased (Figure 1). The maximum adsorption capacity of N and P reached at 155 °C and 160 °C respectively, and the adsorption capacity of P increased significantly when the temperature rose from 150 °C to 160 °C, indicating that certain temperature rise promoted the modification of CS by polyethylene, and resulted in the increase in the adsorption capacity of P. At a temperature below 160 °C, heating increased the strength of SC, while at a temperature over 160 °C, the mechanical strength of SC decreases significantly, indicating that temperature higher than a certain point could seriously damage the structure of SC (Figure 2). The adsorption capacity of P and N on SC increased first and then decreased with the increase of molding time (Figure 1). In general, the adsorption capacity of P and N on SC was significantly different, indicating differences in the adsorption capacity of P and N between CS and SC. Prolonging the molding time to a certain extent will strengthen the adsorption capacity of P and N, while too long molding time will weaken the adsorption capacity of P and N, and greatly weakened the mechanical strength, indicating that longer molding time destroyed the structure of polyethylene and led to a decrease in the mechanical strength of SC (Figure 2).

Researches illustrated that when the mechanical strength was high, biochar could suffer long-term water scouring and air shock (Harpreet & Animesh 2014). At the same time, considering the complexity of water environment and the adsorption capacity of P and N on SC, the best parameters for the preparation of SC were determined.

The Langmuir and Freundlich models were used to fit the adsorption data of P and N on CS and SC. In the Freundlich model, R² < 0.98, while in the Langmuir model, R² > 0.99, showing that the Langmuir model can better describe the adsorption of P and N on CS and SC than the Freundlich model, so the Langmuir model was adopted for analysis. In the Langmuir model, R² > 0.99, it indicated that the distribution of surface adsorption sites of CS and SC tended to be even (Kyung-Won et al. 2015; Qi et al. 2018) (Table 2). The fitted R² value in the Langmuir model on the adsorption capacity of P on CS and SC was higher than that of N, and the fitted R² value in the Langmuir model on the adsorption capacity of P and N on CS was higher than that on SC. This indicated the difference in the distribution of surface adsorption sites between SC and CS (Jiang et al. 2019a, 2019b; Zhang et al. 2019). To determine whether the adsorption of P and N on CS and SC is favorable and spontaneous, the R_L value can be used as a reference, and the formula is expressed as follows:

Where, when R_L = 0, adsorption is an irreversible process; when 0 < R_L < 1, adsorption is a favorable and spontaneous process; when R_L = 1, adsorption is a linear process; when R_L > 1, adsorption cannot be done spontaneously. The R_L value between 0 and 1 indicates that the adsorption of P and N on CS and SC was favorable (Xin et al. 2016) (Table 2). Also, data show that with the increase of c0, R_L value decreases continuously, indicating that the higher the pollutant concentration is, the more it is conducive to adsorption. The Langmuir model is based on the principle of monolayer adsorption and is suitable for chemical adsorption, while the Freundlich model is an empirical one: it illustrates the unevenness of adsorbent surface adsorption sites. Thus, the adsorption of P and N on CS and SC is more inclined to a monolayer adsorption (Ai et al. 2011; Qianqian et al. 2018).

The adsorption process is strongly dependent on the adsorption temperature, and the study of thermodynamics can help to intuitively judge the physicochemical action in the adsorption process. The adsorption process is mainly controlled by such thermodynamic parameters as Gibbs free energy (ΔG), entropy changes (ΔH) and enthalpy changes (ΔS) (Table 3). The values of ΔG in the process adsorption of N and P on CS and SC are all less than 0, indicating that the adsorption process is a spontaneous one (Dilek & Semra 2015), and the higher the temperature is, the lower the values of ΔG are, indicating that the temperature rise is conducive to the adsorption of N and P on CS and SC. With the temperature increase (Figure 3), the adsorption capacity of P and N on CS and SC increased, indicating that the adsorption process was endothermic, and the increase of temperature was conducive to the adsorption process. Moreover, the adsorption capacity of P on CS was greater than that of N on CS, and the adsorption capacity of P and N on CS was more affected by temperature than that on SC. Therefore, the higher the temperature is, the greater the adsorption capacity of N and P is, which is similar to the analysis of RL value. In general, the value of ΔH between 0 and 20 kJ mol⁻¹ indicates the major role of physical adsorption in the adsorption process; when the value of ΔH greater than 0–20 kJ mol⁻¹ indicates the major role of physicochemical adsorption in the adsorption process (Duo et al. 2019). The values of ΔH are both greater than 0–20 kJ mol⁻¹ in the process of adsorption of P on CS and SC, indicating the process of physicochemical adsorption of P. In contrast, the values of ΔH in the process of adsorption of N on CS and SC are both less than 0–20 kJ mol⁻¹, indicating the process of physical adsorption of N. Usually, in the process of liquid phase adsorption, water molecules enter into the adsorbent adsorption sites prior to other substances due to infiltration, the internal system tends to be stable, with a negative value of entropy changes; as adsorbates enter the adsorbents, water molecules are forced to leave adsorption sites to adsorbates; generally the size of water molecules is smaller than adsorbates, thus adsorption of an adsorbate equal to a molecular requires a detachment of several water molecules; thus, the stability of the system is destroyed, leading to an increase in the degree of chaos and the positive value of entropy changes. The stronger the affinity between adsorbents and adsorbates is, the more chaotic the system is and the greater the entropy change is. Both the ΔS values in the process of adsorption on CS and SC are positive; and the ΔS value in the process of adsorption of P on CS and SC is greater than that of N, indicating a certain affinity between P molecules and CS and/or SC. The ΔS value in the process of adsorption of N on SC is greater than that on CS, indicating that polyethylene addition increases the affinity of SC to N (Yiming et al. 2017).

After fitting the data with pseudo-first-order kinetics, pseudo-second-order kinetics, n-order kinetics, and Elovich models, select the one with better fit for research.

The process of adsorption of N and P by SC was kinetically fitted, and the R² values of the pseudo-second-order kinetic model of 0.9963 (N) and 0.9633 (P) were greater than the R² values of the pseudo-first-order kinetic model of 0.9301 (N) and 0.7358 (P), the pseudo-second-order kinetic model has a better fitting effect, which indicates that the process of N and P adsorption by SC is mainly controlled by chemical adsorption (Table 4) (Li et al. 2016; Yang et al. 2018). From the relationship between adsorption amount and time, it can be seen that the adsorption process of N and P by SC is relatively fast before 1 h, the adsorption of N and P is relatively slow between 2 and 4 h, and the adsorption of N and P is close to equilibrium after 6 h (Figure 4). The N equilibrium concentration calculated by pseudo- second-order kinetics is 7.9583 mg·g⁻¹ and the P equilibrium concentration is 9.7078 mg·g⁻¹, which are basically similar to the experimental values of N and P concentration of 7.6856 and 9.6574 mg·g⁻¹.

The pH_pzc value of SC was 8.57 (Figure 5). The adsorption amount of N to SC decreased with the increase of pH value, then increased and then decreased. When the pH value was less than 8, N mainly existed in the form of of NH₄⁺ (Jiang et al. 2019a, 2019b), which repelled the positively charged SC, so the adsorption amount of N was small. When the pH value increases, N exists in the form of NH₃. At this time, the negatively charged SC will attract the non-pointed NH₃. When the pH value is 10, the negative charge on the surface of the SC is larger, and the NH₃ content is higher, and the electrostatic attraction of the two enhanced, the adsorption capacity reached the maximum value. The adsorption capacity of SC to P decreased with the increase of pH value, then increased and then decreased, indicating that the adsorption of SC to P was inhibited under acidic conditions, and the adsorption capacity of P was less affected under alkaline conditions. Relevant studies have shown that when the pH of the solution is different, the existing forms of P are different, mainly H₃PO₄, H₂PO₄⁻, HPO₄²⁻, PO₄³⁻4 forms (Krueger et al. 2021). In the pH value range studied, P mainly exists in two forms, H₂PO₄⁻ and HPO₄²⁻. When the pH value is low, there is a large amount of H⁺ in the solution, which reduces the negatively charged groups on the surface of SC, and the overall performance is electropositive. It is beneficial to the adsorption of positively charged groups. When the pH value increases, the negative charge on the surface of SC increases, which reduces the electrostatic repulsion and increases the adsorption capacity of positively charged groups.

The content of C and H in SC is higher than that in CS, in contrast to a lower content of O in SC than that in CS (Table 5). This may be attribute to the addition of polyethylene (simple structure-[-CH₂-CH₂-]n-), which increases the content of C and H in SC, while the content of O decreases due to the increase in the content of C and H. The lower the content of functional groups containing O, the stronger the hydrophobicity of biochar is (Masebinu et al. 2019). The O/C ratio in SC is much lower than that in CS, indicating that SC is more hydrophobic than CS. The contact angle of SC and CS ranges from 85.70° and from 101.00° respectively (Figure 6). A contact angle less than 90° means that the material is hydrophilic; a contact angle greater than 90° means that the material is hydrophobic; a contact angle of SC close to 90° means that SC has a certain hydrophobicity, which also indicates that SC has a stronger hydrophobicity than CS. Therefore, it can be inferred that the enhancement of SC hydrophobicity is mainly caused by the addition of polyethylene. It is generally believed that the O/C ratio of biochar less than 0.2 is the most stable structure (Kuhlbusch 1995; Spokas 2014). The O/C ratio of SC is 0.17, indicating that SC is highly stable, can exist in the environment for a long time stably, and has a great potential for the treatment of eutrophication water, which is similar to the finding of SC with a mechanical strength greater than 100 N (cm²)⁻¹.

The specific surface area (S_BET), pore volume (Vpore) and micropore volume (V_mic) of SC are significantly lower than those of CS (Table 6), indicating that polyethylene is fully combined with CS. The difference in S_mic between CS and SC is small, since micropores play an important role in the adsorption of pollutants on biochar, it suggests SC still has the ability to absorb pollutants. The values of mean pore size (D_pore) in SC and CS are similar, indicating no significant effect of the addition of polyethylene on the mean pore diameter. In general, compared with CS, the pore structure of SC has a great change.

AFM images of CS and SC are shown in Figure 7. The surface of CS is rather smooth, rather even and loosely distributed, with an average surface roughness of 0.166 μm (Figure 7(CS-a)), while the surface of SC is rough and rich in particles, with an average surface roughness of 0.446 μm (Figure 7(SC-a)). Compared with CS, the surface morphology of SC is significantly changed, indicating the full combination between HDPE and CS. The surface particles of CS biochar are closely bonded, with obvious particles and with roughness and pore diameter increased. CS is rich in a variety of biochar particles with complex and diverse shapes (Figure 7(CS-b)). Although the particle sizes of SC biochar varies greatly, there are fewer types of particle sizes (Figure 7(SC-b)). Compared with CS, the small peaks of SC are significantly reduced while the large peaks are significantly wider, indicating that polyethylene combines with CS biochar particles to form biochar with a certain structure. This conforms to the findings that CS is characterized by a loose surface without an obvious structure (Figure 7(CS-c)), while SC is characterized by an obvious structure of groups (Figure 7(SC-c)).

The specific surface area (S_BET), pore volume (V_pore) and micropore volume (V_mic) of SC are significantly lower than those of CS (Table 6), indicating that adsorption capacity of SC for N P is lower than that of CS. This is the same as that the maximum adsorption (Q_m) of SC to N P is lower than that of CS (3.2 Isothermal adsorption). The SC surface is rough, with an average surface roughness of 0.446 μm (Figure 7(SC-a)), the adsorption capacity of SC, CS to N, P is different (3.3 Adsorption Thermodynamics). The surface morphology of SC,CS is quite different, but the difference in the Qm of N, P between SC, CS is small. Combining the analysis of 3.6. Characteristics, it can be seen that this is because HDPE has played a rode in the adsorption of N, P.

The XRD pattern of CS (Figure 8) shows the diffraction peaks of crystal SiO₂ (PDF#99–0088), the diffraction peaks at 2θ = 26.64 °, 50.14° and 59.96 ° could be assigned to (011), (112) and (121) planes, respectively. While, the XRD pattern of SC shows the diffraction peaks of (CH2)x (PDF#40-1995). The diffraction peaks at 2θ = 21.48°, 23.84° could be assigned to (110), (200)(CH2)x planes, and the diffraction peaks at 2θ = 26.64° could be assigned to (011) SiO2 plane. The weakening of the diffraction peak of SiO₂ crystal in the XRD pattern of SC is mainly caused by the addition of polyethylene, which covers the diffraction peak.

XRD patterns showed the increase in the diffraction peak intensity of SiO₂ after adsorption of N and P on CS, and generally the diffraction peak intensity follows CS-N > CS-P SiO₂, especially the intensity at the peak at 2θ = 26.64° on the (011) SiO₂ crystal plane, which indicates that N and P are bound to the surface of SiO₂, and it is easier for N to bind SiO₂ than P, and the SiO₂ (011) crystal plane is an important N and P binding site (Figure 9). The peak intensity at 2θ = 26.64 °on (011) SiO₂ crystal plane after adsorption of N and P on SC is also enhanced, with CS-N > CS-P, indicating the role of SiO₂ in the adsorption of N and P on SC. After the adsorption of P on SC, the diffraction peak of polyethylene shows no significant difference, indicating a weak affinity between polyethylene and P, which was similar to the research results of Dadong Shao (Kuhlbusch 1995). After adsorption of P on SC, the peak intensity of polyethylene at 2θ = 21.48° and 23.84° on (110) and (200) crystal planes significantly increases, indicating a strong affinity of polyethylene with N.

XPS spectra of CS (Figure 10(CS)) clearly showed peaks at P2p, C1 s, K2p, N1 s and O1 s, in contrast to peaks at P2p, S2p, C1 s, K2p, N1 s, O1 s in XPS spectra of SC. Contents of elements were obtained through calculation of each peak area. The element content in CS is as follows: P (0.37%), C (49.62%), K (7.97%), N (1.1%) and O (40.94%), in contrast to P (0.73%), C (34.64%), K (13.98%), N (1.63%), O (48.16%) and S (0.86%) in SC. The increase of K content from 7.97% to 13.98% may be caused by further exposure of K in SC. The change of C and O content in CS and SC was mainly caused by the addition of polyethylene. Through a further analysis of C1 s and O1 s spectra, 4 peaks occurred in C1 s, which were located at 284.8 eV, 286.31 eV, 289.41 eV and 293.11 eV respectively, representing C-C/C = C/C-H, C-O, CO32- and COOH (the K functional group at 295.9 eV) respectively (Spokas 2014; Shao et al. 2017); and 4 peaks occurred in O1 s, located at 531.1 eV, 532.6 eV, 533.95 eV and 534.08 eV respectively, representing C = O, C-O, O = C-O, COOH (Chen et al. 2021; Fan et al. 2021). Compared with CS, in terms of C functional group, C-C/C = C/C-H peak area of SC decreased and COOH increased; while in terms of O functional group, C-O peak area of SC significantly decreased in contrast to an increase in O = C-O and C = O in SC. This may be caused by the addition polyethylene. In light of the molecular structure of polyethylene, it is speculated that polyethylene binds CS biochar particles in the form of adhesion, coating and encapsulation, resulting in the coverage and exposure of these functional groups.

The FTIR spectra showed 7 major peaks (at 687 cm⁻¹, 1,002 cm⁻¹, 1,463 cm⁻¹, 2,356 cm⁻¹, 3,126 cm⁻¹, 3,610 cm⁻¹, 3,732 cm⁻¹) on CS. After adsorption of P, the peaks at 1,463 cm⁻¹, 2,356 cm⁻¹, and 3,126 cm⁻¹ significantly decreased, with functional groups C = C, C-H, O-H respectively. This indicates the decrease in the intensity of these 3 peaks due to the adsorption of P. Peak intensity of functional group C = C decreased and the absorption peak shifted to high wave numbers, which was affected by negative charge of P electrons. The distribution of electrons changed through the electrostatic induction, with the increase in vibration frequency of the group function C = C, and the shift to high wave numbers as a result. Peak intensity of functional group O-H decreased and shifted to high wave numbers, because the negative charge of O electrons was higher than P after the adsorption of P, resulting in the shift of H to O. Also, after adsorption of N, the intensity peak at 1,463 cm⁻¹, 2,356 cm⁻¹ and 3,126 cm⁻¹ significantly decreased, with a lower intensity at 1,463 cm⁻¹ and 2,356 cm⁻¹ than that of P, which was due to the difference in the negative charge of N and P electrons.

FTIR spectra showed 9 major peaks on SC (mainly at 703 cm⁻¹, 1,004 cm⁻¹, 1,407 cm⁻¹, 1,463 cm⁻¹, 2,357 cm⁻¹, 2,848 cm⁻¹, 2,920 cm⁻¹, 3,610 cm⁻¹, 3,741 cm⁻¹) (Figure 11). After adsorption of P, the intensity peaks changed at 704 cm⁻¹, 1,004 cm⁻¹, 1,407 cm⁻¹, 1,463 cm⁻¹, 1,463–2,357 cm⁻¹, and 920–3,610 cm⁻¹. The peak at 704 cm⁻¹ and 1,004 cm⁻¹ showed an asymmetric stretching vibration of S-H bond and P-O bond. Generally, atoms N, S, O and P have higher negative charges, and will form hydrogen bonds with H atom. As the negative charge of P is higher than that of S, thus such a hydrogen bond as S-H-P with a bias toward P is formed (Honghong et al. 2017; Qi et al. 2018). The peak at 1,407 cm⁻¹ and 1,463 cm⁻¹ showed the stretching vibration of C = C (Gang et al. 2016). After adsorption of P, electron distribution changed due to the influence of the negative charge of P through electrostatic induction, making the absorption peak of C = C shift to the high wave number. The peak at 2,920–3,610 cm⁻¹ showed the stretching vibration of -OH. After adsorption of P, the peak strength weakens to a certain extent, indicating the formation of such hydrogen bonds as O-H-P. The changes in the peaks at 703 cm⁻¹, 1,004 cm⁻¹, 1,407 cm⁻¹, 1,463 cm⁻¹, 1,463–2,357 cm⁻¹ and 2,920–3,610 cm⁻¹ showed a similar trend after adsorption of N to that of P, however, the peak intensity at 1,004 cm⁻¹, 1,407 cm⁻¹ and 1,463 cm⁻¹ was slightly lower than that of P. This is similar to the peak intensity change at 1,463 cm⁻¹ and 2,356 cm⁻¹ after adsorption of P and N on CS.

Thus, the adsorption of P on SC is a physicochemical adsorption process, and P can be attached to the internal and external pores of SC; the functional groups C = C and O-H can be attracted by static electricity with P. The functional groups S-H and O-H can form hydrogen bonds with P. The adsorption of N on SC is a physicochemical adsorption process with the dominance of physical process, and N can be filled in internal and external pores of SC. The functional groups C = C and O-H can be attracted by static electricity with N. The functional groups S-H and O-H can form hydrogen bonds with N. In conclusion, the adsorption mechanism of P and N on SC can be attributed to pore filling, electrostatic attraction and hydrogen bonding.

SiO₂ crystal is the main crystalline substance in CS. Significant changes in XRD spectra can be observed through the comparison of CS before and after adsorption of N and P, mainly on the 26.64° (011) SiO2 crystal plane, which indicated the impact of N and P loadings on the crystal structure. Federico (Musso et al. 2011) calculated that the electronic interaction energies of SiO₂ (001,100,011) were −56.4, −64.9 and −77.5 kJ mol⁻¹ and −46.4, −65.4 and −84.7 kJ mol⁻¹ for silica – H₂O system and silica – NH3 system respectively, and reported the electron binding energy was the main energy in the final interaction energy. The great changes of crystal plane 011 in.XRD spectra of CS before and after adsorption of N and P illustrated that SiO₂ crystal had an electrostatic bond with N and P. XRD spectra showed that SC was rich in polyethylene and SiO₂, and the diffraction peak of polyethylene changed significantly after adsorption of N and P, indicating that the attachment of N and P changed the structure of polyethylene. (Shao et al. 2017) conducted XPS characterization of phosphate functionalized polyethylene (PO₄/PE) via H₃PO₄ treatment and found that C-O-P was contained in PO₄/PE. Considering that HDPE is rich in C-O-H, it can be speculated that P can be bound to HDPE in the way of C-O-H → C-O-P. Meanwhile, because of the similarity of HDPE changes in XRD spectra of SC after adsorption of N and P, it can be inferred that N could be bound to HDPE in a same way.

The XPS and FTIR spectra showed that CS and SC surface were rich in functional groups. The FTIR spectra of CS before and after adsorption of N P showed significant changes in peaks at 1,463 cm⁻¹, 2,356 cm⁻¹ and 3,126 cm⁻¹, indicating that C = C, C-H and O-H functional groups all played a role. The peak intensity of C = C weakened and shifted to high wave number, indicating the impacts of N and P on the electron distribution of C = C. Since P is more electronegative than N, P can be bound with C = C more easily; consequently, the peak intensity of C = C in CS-P was much weaker. The decrease in the peak intensity of O-H and its shift to high wave number were similar to that of C = C. (Yongwei et al. 2019) also reported the role of hydroxyl groups in the adsorption of organic phosphorus (OP) and inorganic phosphorus (IP) on modified corn bracts. (Rahman et al. 2021) illustrated that the presence of a P-O-CH3 group gave rise to an intensive absorption bang around the 1,040 cm⁻¹ through the FTIR characterization of modified biochar. The peak intensity of CS around 1,002 cm⁻¹ was significantly enhanced before and after adsorption of N and P, indicating that O-H adsorbed P through O-P. After the adsorption of P on SC, the attenuation of S-H (704 cm⁻¹) absorption peak was due to the formation of hydrogen bonds like S-H-P, and the attenuation of the -OH peak (2,920–3,610 cm⁻¹) was due to the formation of hydrogen bonds such as O-H-P. After the adsorption of N on SC, the attenuation of the absorption peak at 704 cm⁻¹ was due to the formation of hydrogen bonds such as S-H-N, and the attenuation of absorption peak at 2,920–3,610 cm⁻¹ was due to the formation of hydrogen bonds such as O-H-N.

According to the analysis of isothermal adsorption and adsorption thermodynamics, the adsorption of P on SC is a physicochemical adsorption process, by contrast, the adsorption of N on SC is a physically dominated physicochemical adsorption process. N and P can be attached to the internal and external pores of SC, and be loaded on SiO₂ and HDPE; C = C, O-H functional groups can generate electrostatic attraction with N and P; S-H and O-H functional groups can form hydrogen bonds with N and P. In conclusion (Figure 12, Figure 13), the adsorption mechanism of P and N on SC can be attributed to pore filling, electrostatic attraction and hydrogen bonding.

Preparation of biochar with lotus-like structure (Figure 14): the composite of HDPE with 60% seizing amount and CS is molded in a lotus-like silastic molder with an inner diameter 5 cm, an outer diameter 8.5 cm and height 4 cm under the pressure of 5 N (cm²)⁻¹, heated at 160 °C for 95 min, and finalized after cooling. Lotus-like biochar can float on the water, adsorb nutrients such as N and P in the water, provide nutrients for aquatic plants and space for the growth of micro-organisms, and form a waterscape in combination with others.

Preparation of hinged biochar (Figure 15): the composite of HDPE with 60% seizing amount and CS is molded in a hinged biochar brick molder under the pressure of 5 N (cm²)⁻¹, heated at 160 °C for 95 min, and finalized after cooling. Hinged biochar can adsorb such nutrients as N and P in the water, provide nutrients for the growth of aquatic plants and organisms. The stabilized hinged structure can conserve water and soil, planting pores with different sizes can meet the growth demand of plants, and form a plant landscape in combination with others.

In this study, with the consideration of the adsorption capacity of N and P on SC via different processes and mechanical strength greater than 100 N (cm²)⁻¹, the optimal parameters for SC preparation was determined as follows: the composite with HDPE (200 mesh) mixed with CS at a mass ratio of 3:2 was first molded under a pressure of 5 N (cm²)⁻¹, heated at 160 °C for 95 min, and finalized after cooling. The adsorption of N and P on SC conforms to the Langmuir model. The distribution of adsorption sites on SC surface tends to be even, and the adsorption tends to be monolayer adsorption, with the main role of chemical adsorption. The RL value indicates that the adsorption of N P on SC is favorable and spontaneous, and the temperature rise will promote the adsorption of N and P on SC. The adsorption process of P on SC is a physicochemical adsorption process, while the adsorption process of N on SC is a physical adsorption process. The affinity between P and SC is higher than that of N. The data fitted by the pseudo-second-order kinetic model for the adsorption of N and P by SC has a high degree of fit, indicating that the adsorption of N and P by SC is mainly controlled by chemical control; the equilibrium concentration of N calculated by the pseudo-second-order kinetic model is 7.9583 mg g⁻¹, the equilibrium concentration of P is 9.7078 mg g⁻¹. The pH_pzc value of SC was 8.57. The addition of HDPE improves the hydrophobicity and stability of SC, and enhances the roughness of CS particles by bonding them together. According to XPS, XRD and FTIR analysis of SC, N and P are attached to the internal and external pores of SC and can be loaded on SiO2 and HDPE, C = C and O-H functional groups can generate electrostatic attraction with N and P, and S-H and O-H functional groups can generate hydrogen bonds with N and P. Therefore, Figure 12, Figure 13 the adsorption mechanism of P and N on SC can be attributed to pore filling, electrostatic attraction and hydrogen bonds.

This work was supported financially by National Natural Science-Foundation of China (41761098,21767027).

Investigation, Writing-original draft, Formal analysis, Visualization, Software, Methodology, L.F.; Writing-review and editing,L.Y.; Conceptualization, Methodology, Writing-review and editing, Supervision, Data curation, W.Y.; Supervision, Visualization, Software,Y.S.; Visualization, Methodology, M.R.

This research was funded by the National Natural Science Foundation of China (41761098,21767027).

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper

All relevant data are included in the paper or its Supplementary Information.