Production and modification of hydrochar from anaerobically digested cattail for adsorbing ammonium and phosphorous in wastewater

Water contamination by nitrogen and phosphorus ions has a direct consequence of eutrophication to the ecosystem. The objective of this study is to investigate the production of hydrochars by acetic acid or sodium hydroxide assisted hydrothermal carbonization (HTC), various activation methods, and the potential of hydrochar as an adsorbent to remove NH4 þ-N and PO4 3 -P from wastewater. The results showed that acetic acid or sodium hydroxide assisted HTC and activation with magnesium chloride or air could improve the surface properties of hydrochar. Acetic acid modification generated extensive oxygenated functional groups, while sodium hydroxide modification produced hydrochar with a high N/C ratio and surface aromaticity. Treating hydrochar with magnesium chloride could impregnate nano-magnesium particles on the hydrochar, thereby improving the ability to remove N and P. Air activation of hydrochar resulted in more oxygen containing functional groups. The NH4 þ-N and PO4 3 -P adsorption capacities of these hydrochars ranged from 92.6 to 122.4 mg/g and 1.6 to 15.8 mg/g, respectively. The adsorption capacity of hydrochars in swine wastewater is similar to the results of artificial wastewater. The results suggested that Mgnanoparticle dispersion and oxygen-containing functional groups played a major role in adsorption than ion exchange and physisorption.


GRAPHICAL ABSTRACT INTRODUCTION
Although the specific adsorption mechanism is still unclear, the presence of functional groups and metals may play a key role in the adsorption of N and P (Takaya et al. 2016).
In the past few years, hydrothermal carbonization (HTC) has gained a considerable attention for producing hydrochar (a typical biochar) from carbonaceous materials with a high moisture content (such as municipal sludge, manure, algae, and kitchen waste) (Zhang et al. 2020). HTC processes the feedstock under a self-generated pressure at a mild temperature (e.g., 250°C) to produce solid hydrochar and aqueous solution containing organic and inorganic compounds. Also, HTC can destroy pathogens, eliminate odors, and remove heavy metals from raw materials. The product of hydrochar is a more hydrophobic solid than that produced by dry torrefaction, which in turn has positive correlation with carboxylic groups on the biochar surface, specific surface area, and pore volume (Xiu et al. 2017;Mao et al. 2019). It has been reported that various factors such as the cation exchange capacity, oxygen containing/acidic functional groups, available anion exchange sites and surface charges, and initial concentration of pollutants affect the adsorption of N and P over hydrochar (Kameyama et al. 2012;Chintala et al. 2014).
In this study, cattail (Typha species) is used as the feedstock, which is a kind of aquatic plant commonly grown on wetlands. It has high insect resistance, high adaptability and high reproduction ability (Zhang et al. 2012). The annual productivity of cattails is considerably high. It was reported that standing crop yields up to 22.4 Mg ha À1 of leaves and 30.9 Mg ha À1 of rhizomes (Dubbe et al. 1988). Improper management of cattails may cause corruption in water, severe secondary pollution, rapid marshes of lakes, and reduced invertebrate production and migratory birds (Kostecke 2002). Studies have shown that harvesting half of the cattails grown in wetlands every year (including rhizomes) will not affect its growth and progress next year (Bansal et al. 2019). Because of these advantages, cattails are considered as a potential renewable feedstock for bioenergy production. Glacial acetic acid (CH 3 COOH) or sodium hydroxide (NaOH) were used to assist the HTC process of cattails. Both acetic acid and sodium hydroxide have been shown to accelerate the degradation/hydrolysis of the dairy manure and improve the biochar's quality (Chen et al. 2018). The obtained hydrochar was further activated by using MgCl 2 or air. To the best of our knowledge, this is the first study on the influence of chemical-assisted HTC of the anaerobic digestate and subsequent MgCl 2 or air activation on the N and P adsorption capacity of hydrochar. The objective of this study is to (1) compare the N and P adsorption capacity of hydrochar produced under different HTC temperatures, chemical modification and MgCl 2 or air activation; (2) study the N and P adsorption capacity of various hydrochars prepared from anaerobic digested cattail in artificial wastewater and swine wastewater. The research results provide a theoretical and practical basis for the development of hydrochar adsorbents for N and P adsorption from wastewater.

Materials
Cattail (Typha latifolia), the anaerobic digestion (AD) process, and the composition of cattail digestate have been described elsewhere (Zhang et al. 2020). Cattail was anaerobically digested for 14 days, collected after AD, washed with deionized water to remove loose contents, dried at 105°C for 24 h in a Thermo Scientific Precision Compact Oven (Waltham, MA, USA). About 2 kg (dry basis) of anaerobically digested cattails were prepared and stored at room temperature for future use.
Artificial wastewater was prepared according to the procedure reported in the literature (Fang et al. 2014). For P-containing wastewater, potassium phosphate monobasic was used for the preparation of the phosphate solution at 125 mg/L. Ammonium hydroxide was used to prepare the N-containing wastewater at a concentration of at 1,000 mg/L. These two solutions were diluted to various concentrations for the measurement of adsorption isotherms.
Swine wastewater was collected from the uncovered lagoon at the swine research facility of the University, which represents a low-strength wastewater. Pretreatment of swine wastewater was carried out by sedimentation and filtration with a Whatman quantitative filter paper with 8 μm pore size to remove large, non-soluble particulate solids.

Hydrothermal carbonization
Approximately 10 g of dried feedstock was mixed with 160 mL of deionized water into a 300 ml stirred Parr reactor (Parr Instrument, Moline, IL, USA). For chemical-assisted HTC, glacial acetic acid (CH 3 COOH) or sodium hydroxide (NaOH) was added with 160 mL water, giving a concentration of 0.6 M and 2 M, respectively. The choice of the concentrations is based on the results reported in the literature (Reza et al. 2015;Wang et al. 2017).
The temperature of the reactor was increased at a heating rate of about 10°C/min and held at the desired temperature for 1 h. HTC was performed at three different temperatures of 200, 250 and 300°C and the stirring speed was set at 100 rpm. The hydrochar produced via water-only, acetic acid-assisted, and sodium hydroxide-assisted HTC at various temperatures was denoted as B200/B250/B300, AB200/AB250/AB300, and NB200/NB250/NB300, respectively.
After HTC treatment, the reactor was cooled down to the room temperature. The hydrochar was filtered out using Whatman grade 1 filter paper and washed with deionized water and ethanol to remove residual bio-oil. The hydrochar was then dried in an oven at 105°C for 24 h and ground into powder for further analysis. The yield of hydrochar was calculated as: Hydrochar yield ¼ Mass of hydrochar Mass of raw biomass Ã100% (1)

Activation of hydrochar
Molten salt activation of hydrochar was prepared with MgCl 2 , which is one of the common molten salts for activation. For a typical process, the hydrochar (B250) was mixed with MgCl 2 at a mass ratio of 1:1, and then the mixture was homogenized with a mortar and pestle. The powdered mixture was transferred to a ceramic crucible and was placed in an electric furnace equipped with a continuous nitrogen flow. After flushing with nitrogen for 30 min, the furnace was ramped at 10°C/min to the carbonization temperature of 600, 700 or 800°C and kept at this temperature for 1 h. The furnace was cooled to ambient temperature by switching off the power; meanwhile, the nitrogen flow was maintained until the temperature reached room temperature. The as-obtained product was washed with sufficient amount of water to remove the salts. The activated hydrochar was dried in an oven at 105°C for 12 h. Hydrochars carbonized at 600, 700 and 800°C were denoted as MB600, MB700, and MB800, respectively. The yields of hydrochars defined by Equation (1) are reported in the Results and Discussion section. Air activation of the hydrochar (B250) was conducted at 250°C. For a typical run, 5 g of hydrochar was heated from the room temperature to activation temperature of 250°C in an air stream. The sample was held at 250°C for 2 hr. Then, under the same flow of air, the system was allowed to cool down to the room temperature. The hydrochar was designated as air activated hydrochar.

Characterization
Physical nitrogen adsorption, pore volume, and pore size were measured using a Micromeritics ASAP 2020 surface area and porosity analyzer (Norcross, GA, USA).
Thermogravimetric analyses (TGA) were performed using a TA Instruments SDT Q600 thermogravimetric analyzer (New Castle, DE, USA), in which samples were heated to 250°C at a heating rate of 20°C/min in nitrogen.
The morphology of hydrochars was evaluated using a JEOL 7500F scanning electron microscope (SEM, Tokyo, Japan). Energy-dispersive X-ray spectroscopy (EDX) analysis of hydrochar was performed on a JEOL 7500F SEM coupled with an energy dispersive X-ray detector. For each sample (total 13 different hydrochars), the measurement was carried out on 5-10 different particles.
The absorption spectroscopy in the infrared region was measured to determine the frequency of vibrations of the functional groups present in the hydrochar. The spectrum readings were obtained using an Agilent 670 Fourier transform infrared (FTIR) spectrometer with attenuated total reflection (ATR) varied in the range of 4,000-400 cm À1 (Santa Clara, CA, USA).

Ammonium and phosphate adsorption
To evaluate the adsorption capacity of hydrochar to ammonium, 0.1 g of hydrochar was mixed with 100 mL ammonium hydroxide solution in a beaker. The mixture was then shaken on a shaker at 160 rpm and 25°C for 24 h. After that, the mixture was filtered through a 0.45 μm membrane, and then the ammonium concentration was determined using the TNT 830 ammonia kit with a Hach 6000 UV/VIS spectrophotometer (Loveland, CO, USA).
The same procedure was used to measure the adsorption capacity of hydrochar to phosphate using 125 mg/L phosphate solution. Phosphate concentration of the filtrate was determined using the TNT 845 kit with the Hach spectrophotometer. The adsorption experiments were performed in triplicate, and the average values were reported.
The adsorption capacity Q (mg g À1 ) of the hydrochar was calculated as: where C e and C 0 are the equilibrium and initial NH 4 þ -N or PO 4 3À -P concentrations (mg/L), respectively. V is the volume of the solution and m is the weight of hydrochar (g). For the adsorption test in swine wastewater, 0.1 g of hydrochar was mixed with 100 mL diluted swine wastewater in a beaker. The mixture was then shaken on a shaker at 160 rpm and 25°C for 24 h. After that, the mixture was filtered through a 0.45-μm membrane and analyzed for the chemical oxygen demand (COD), NH 4 þ -N, and total phosphorus (TP).

Adsorption isotherms and mathematical models
To measure the adsorption isotherm of ammonium, 0.1 g of the hydrochar was mixed with 50 mL ammonium hydroxide solution with different initial concentrations of 25, 50, 100, 200, 300, 500, 750, and 1,000 mg/L. All samples were shaken on the mechanical shaker at 160 rpm and 25°C for 24 h. The samples were filtered through a 0.45-μm membrane, and then the ammonium concentration was determined immediately with the Hach spectrophotometer. Based on the difference between the initial and final solution concentrations, ammonium adsorbed on the hydrochar was calculated. To measure the adsorption isotherm of phosphorus, 0.1 g of the hydrochar was mixed with 50 mL phosphate solution with initial concentrations of 20, 50, 80, 100, 150, 200, 250, 300, and 350 mg/L. The remaining procedure was the same as that of ammonium. All isotherm experiments were performed in triplicate, and the average values reported. The Langmuir and Freundlich models were used for the regression of the adsorption isotherm. The equations of these two models are written as: where K L and K f are the Langmuir bonding term related to interaction energies (L/mg) and the Freundlich affinity coefficient (mg (1Àn) L n /g), respectively. S max is the Langmuir maximum capacity (mg/kg), Ce is the equilibrium solution concentration (mg/L) of the sorbate, and n is the Freundlich linearity constant.

RESULTS AND DISCUSSION
Hydrochar production and activation HTC of anaerobically digested cattails was conducted in deionized water, acetic acid solution and sodium hydroxide solution at 200, 250 and 300°C for 1 h. The hydrochar yields are summarized in Figure 1. The reaction temperature showed an obvious effect on the hydrochar yield. When the temperature increased from 200 to 300°C, the hydrochar yields decreased  On average, the increase of the temperature from 200°C to 300°C reduced the solid yield by 7-9% of the original mass. The loss of biomass during HTC at a low temperature was mainly attributed to the decomposition of the hemicellulose component. As a result, the hydrochar product treated at 200°C still showed a color similar to the original feedstock. When the temperature increased, cellulose degradation starts to occur at a temperature above 200°C due to disintegration through hydrolytic action and further reactions, decreasing the hydrochar yield (Volpe et al. 2018). Because of the p-value of one-way ANOVA analysis (Table S1), the application of acetic acid and sodium hydroxide significantly decreased the hydrochar yield at 250-300°C (i.e., p-value , 0.05) compared with water-only HTC, which may be caused by the higher hydrolytic effect caused by chemicals added. It is well known that acetic acid could enhance the degradation of hemicellulose and cellulose, while sodium hydroxide could dissolve lignin and partial hemicellulose .
Because raw hydrochars and acetic acid/sodium hydroxide modified hydrochars showed similar properties and adsorption capacity, B250 hydrochar was chosen for further activation. MgCl 2 activation of B250 hydrochar was conducted at 600, 700 and 800°C, and the yield of activated hydrochar was approximately 90% of the original hydrochar. Air activation process was simulated by using TGA. Figure 2 shows the TGA-DSC profiles of B250 hydrochar activated in air. When the temperature increased from the ambient temperature to 250°C, the weight loss can be divided into two phases (Figure 2(a)). The first phase was between the ambient temperature and 150°C, which was due to the loss of water present in various forms and represented about 3.5 wt% of the total hydrochar weight. At this stage, the process was endothermic. The second phase happened between 200 and 250°C, representing a weight loss of approximately 0.5 wt%. The weight loss in the second phase might be due to volatilization of organic matters or oxidation. If the hydrochar was held at 250°C for 2 h, it would cause an additional 20% mass loss. The process became exothermic (Figure 2(b)).

Characterization of hydrochar
Surface area Table 1 gives the surface area of HTC hydrochars and activated hydrochars. The BET surface area of HTC hydrochars increased with the HTC temperature. The HTC hydrochar produced with deionized water at 200, 250 and 300°C had a relatively small surface area of 4.40, 4.95 and 8.96 m 2 /g, respectively. These values matched the results reported in the literature (Jaruwat et al. 2018). Hydrochars produced via acetic acid and sodium hydroxide-assisted HTC showed a similar trend. According to the p-value of one-way ANOVA analysis (Table S2), the surface area of hydrochars obtained at 300°C was significantly higher than that of hydrochars obtained at 200-250°C. When the temperature increased from 200 to 300°C, the BET surface areas of the hydrochars increased from 2.52 to 6.86 m 2 /g and 4.71 to 7.42 m 2 /g for acetic acid and sodium hydroxide treated samples, respectively. This trend might be caused by the gradual degradation of organic matter in the feedstock and the formation of new vascular bundle and channel structure during the HTC process. But the difference of the surface area of hydrochars obtained via water-only hydrothermal liquefaction (HTL), acid-assisted HTL and base-assisted HTL was not significant.
Compared with HTC hydrochars, hydrochars activated with MgCl 2 showed higher BET surface areas, which were 84.85 + 1.22, 131.53 + 2.47, 151.02 + 2.86 m 2 /g for 600, 700, and 800°C treated samples, respectively. Nano-sized MgO dispersion on the hydrochar surface may increase the specific surface area of Mg-loaded hydrochars, which is further illustrated by the SEM images and the magnesium content obtained by the elemental analysis. The hydrochar activated in air at 250°C showed the highest BET surface area (337.75 m 2 /g) among all modified hydrochars. Activation in the presence of air leads to the development of porosity, mostly due to the generation of micropores and mesopores (Sajjadi et al. 2019).
Scanning electron microscopy and EDX analysis SEM images of hydrochars are shown in the supplementary material ( Figure S1). SEM analysis of hydrochars allows the visualization of the surface characteristics of the materials. In hydrochars produced via HTC at 200-300°C, the appearance of a rough surface and a rupture into scales were observed, which may be due to the intrinsic nature of the hydrochar matrix. Chemical modification with acetic acid and sodium hydroxide resulted in considerable cracks and fine particles attached to the surface, forming a system of micro-pore networks. In addition, the sodium hydroxide-modified hydrochar showed sodium hydroxide crystals on the surface, and the length of the crystals was between 1-2 μm.
The morphology of magnesium chloride-treated hydrochars is largely dependent on the treatment temperature. When treating at 600 and 700°C, a large number of MgCl 2 crystal structures emerged and covered the surface. When a treatment temperature of 800°C was used, molten salt activation happened, and the molten MgCl 2 corroded the carbonaceous material to form more cracks and voids, resulting in a higher surface area. These finding are backed up by similar results reported by Fang et al. (2014) in magnesium-modified corn hydrochar. Air-activated hydrochar showed more cracks and carbon particles on the surface compared with B250 hydrochar. Table 2 gives the elemental compositions of the raw and modified hydrochars determined by EDX. The dominant elements were carbon and oxygen, which were from 59.98% to 82.49% and 11.35% to 32.28%, respectively. Generally, the hydrochar obtained at 200°C had the lowest carbon content, indicating incomplete volatilization. Hydrochars produced at 250 and 300°C showed a similar quality. Because a 250°C HTC could give a higher hydrochar yield, 250°C was considered as the optimal HTC temperature in this study. Therefore, hydrochar obtained via the 250°C HTC was further used for activation.
The hydrochars also had trace amounts of phosphorus, magnesium, potassium, calcium, and sulfur. Sodium hydroxideassisted HTC yielded hydrochar with a higher amount of sodium, which is the residue of sodium hydroxide. The presence of magnesium contents on magnesium modified hydrochar is related to the amount of MgCl 2 used and the carbonization temperature. The loading of magnesium was between 3.64 and 4.02%. Carbonization at 800°C considerably reduced the chlorine content of the material, while at 600°C or 700°C, magnesium still existed in the form of MgCl 2 . These results were further corroborated by the dispersion of Mg on the surface and enhanced surface area. The oxygen content of air activated hydrochar increased at the expense of the carbon content, which is attributed to the oxidation of the material.

FTIR analysis
The analysis of functional groups on the surface of hydrochars could help to determine the mechanisms responsible for the adsorption of nutrients in the structure of these materials. The infrared spectra of hydrochars are shown in Figure 3. The FTIR analysis demonstrated that the character peaks of modified hydrochars differed from each other. The peak at approximately 3,445 cm À1 corresponded to the O-H stretching vibration, which changed noticeably for MgCl 2 and acetic acid modified samples. The weakened peak of the O-H vibration indicates the structural change of the feedstock under these chemical treatments. Peaks obtained between 2,820 and 2,990 cm À1 suggested the stretching vibrations of aliphatic C-H. This peak indicated the presence of aliphatic structure in hydrochar samples. The band at 1,737 cm À1 indicated the C ¼ O stretching vibrations in all modified hydrochar, suggesting hemicellulose degradation during HTC above 200°C. The peak at 1,698 cm À1 corresponding to the carboxyl, ester or carbonyl groups was seen in all hydrochars with little difference, indicating the presence of oxygen-containing functional groups on the hydrochar surface. However, this peak was not clearly noticeable in magnesium-modified hydrochar at 800°C, which may be due to increased oxygen loss caused by thermal degradation and decarboxylation (Reza et al. 2015). This loss of oxygen in magnesium-modified hydrochar at higher temperature is supported by elemental analysis as well. The adsorption peak at 1,508 and 1,460 cm À1 indicated the presence of aromatic lignin groups in the sample. However, the flattened peak at higher temperature magnesium-modified hydrochar suggested the partial degradation of lignin.
Overall, magnesium modification did not affect the structural formation of most organic functional groups in Mg/hydrochar. The character peaks at 3,200 cm À1 -3,700 cm À1 , 1,400 cm À1 -1,690 cm À1 , and 550 cm À1 -850 cm À1 represented the hydroxyl (O-H), double bonds (C ¼ O and C ¼ C), and organic chloride (C-Cl), respectively. Air activation reduced certain structures of the hydrochar. But air activated hydrochar was still rich in functional groups of O-H, C ¼ C, carbonyl group, C-O, and so on.

NH 4 þ -N and PO 4 3À -P adsorption
Factors including initial NH 4 þ concentration, pH, and contact time with solution could determine the adsorption capacity of hydrochar. In this study, the pH of the solution was adjusted to neutral as it has been shown to favor NH 4 þ -N adsorption (Banerjee et al. 2014). As shown in Table 3, hydrochar produced by HTC resulted in around 10% of NH 4 þ -N removal from the artificial wastewater. Hydrochar produced by water-only and acetic acid-assisted HTC resulted in a PO 4 3À -P removal of 2-3%. Sodium hydroxide modification of the hydrochar at 250 and 300°C increased the PO 4 3À -P adsorption to ∼4.5%.
Increased surface area and addition of functional groups may contribute to the increased adsorption capacity.  Magnesium chloride modification of the hydrochar (B250) increased the adsorption capacity of both NH 4 þ -N and PO 4 3À -P to ∼12% and ∼9.5%, respectively. Compared with the B250 hydrochar, the adsorption capacity of MgCl 2 -activated hydrochar increased by 20 and 240%, respectively. The NH 4 þ -N adsorption capacity of hydrochar prepared from AD digestate is similar or higher than the results reported in the literature, while the PO 4 3À -P adsorption capacity is lower than the results reported in the literature. Takaya et al. (2016) reported that NH 4 þ -N and PO 4 3À -P adsorption capacities of hydrochar ranged from about 105.8-146.4 mg g À1 and 0 to 30 mg g À1 , respectively. Another study showed that the cassava straw and banana straw biochars exhibited the theoretical maximum saturated adsorption capacities of 24.04 mg·g À1 (NH 4 þ -N) and 31.15 mg·g À1 (TP), respectively (Jiang et al. 2019). They suggested that the magnesium loading had strong correlation with NH 4 þ -N and PO 4 3À -P adsorption. In this study, surface area and pore volume showed moderate and weak correlation for NH 4 þ -N and PO 4 3À -P adsorption, respectively. The improved NH 4 þ -N adsorption capacity may be due to the presence of increased negativelycharged functional groups like carboxyl group. Increased PO 4 3À -P adsorption capacity could be attributed to surface complexation with hydroxyl functional group and Mg-phosphate precipitation reaction ( Jiang et al. 2019). Air-activated hydrochar had the most influential impact on P adsorption as it resulted in 15.8 mg/g of P adsorption and 12.6% removal from wastewater. Meanwhile, it also showed a slightly higher NH 4 þ -N adsorption capacity. High surface area of air activated hydrochar (337.75 m 2 /g) may have allowed high adsorption of both NH 4 þ -N and PO 4 3À -P. Additionally, the increase in negative surface charge due to the presence of acidic functional groups such as the carbonyl group in air-activated hydrochar may have favored the N sorption capacity.

Adsorption capacity in swine wastewater
Swine wastewater was used to test the adsorption capacity of hydrochars, which contained 1,470 mg/L COD, 50 mg/L NH 4 þ -N and 151 mg/L TP. The phosphorus adsorption capacity of hydrochars in swine wastewater is similar to the results of artificial wastewater, which is within a range of 1.5-5.8 mg/g (Table 4). Because the swine wastewater collected from the open lagoon contained a small amount of NH 4 þ -N, almost 100% of NH 4 þ -N was removed by hydrochars. However, with the exception of air-activated hydrochar, most hydrochars had leaching problems. Therefore, before using hydrochar or activated carbon as an adsorbent to treat wastewater, pretreatment is required to reduce leaching.

Ammonia and phosphate adsorption isotherms
Based on the initial evaluation, the following five hydrochars were selected for the adsorption isotherm study, including B250, AB250, NB250, MB800, and air-activated hydrochar. The results were fitted into Equations (2) and (3). tion, the increase in initial concentration up to 300 mg/L showed a sharp increase in the adsorption capacity by hydrochars and stabilized at higher initial concentration. The PO 4 3À adsorption showed a similar pattern, in which there was a rapid increase of adsorption capacity with the increase of initial concentration up to 150 mg/L and stabilization at higher initial concentration. Adsorption equilibrium concentration was reached after 24 h at room temperature for both NH 4 þ -N and PO 4 3À -P. Adsorption equilibrium may have resulted in such phenomena. Compared with PO 4 3À -P, overall higher adsorption of NH 4 þ -N might be due to the intrinsic negative charge of hydrochar that is not conducive for P adsorption.
The isotherm parameters for both models with the linear regression coefficient (R 2 ) are presented in Table 5. The R 2 value suggested that the Freundlich model fits the NH 4 þ -N adsorption data better than the Langmuir model for most types of hydrochar. Typically, the Freundlich model indicates heterogenous and multilayer adsorption. Therefore, the results suggest that adsorption could be described by heterogenous and multilayer adsorption. This in turn implies that N adsorption by hydrochar is a sum of all the adsorption sites, where stronger binding sites are occupied first. The maximum adsorption capacity of NH 4 þ -N by B250, AB250, NB250, MB800, and air-activated hydrochar were 147.08, 100, 98.04, 98.04, and 90.90 mg/g, respectively. The results showed that although the adsorption capacity increased at the initial NH 4 þ -N concentration, the saturation of adsorption sites during the isotherm adsorption process led to a decrease in the adsorption capacity at a higher initial concentration. The maximum NH 4 þ -N adsorption capacity of hydrochars has the following order: B250 . AB250 .
NB250 ¼ MB800. air-activated hydrochar. From Table 5, the R 2 value indicates that the results of PO 4 3À -P adsorption also fit the Freundlich model better than the Langmuir model. The Freundlich model suggests that the adsorption of P is heterogeneous and multilayer, which is due to the diversity of adsorption sites and the nature of ions as free or hydrolyzed substances. The literature also showed that the Freundlich model suits phosphate adsorption better because of saturated adsorbents or precipitation reactions (Zeng et al. 2013). The maximum PO 4 3À -P adsorption capacity of hydrochars has the following order: air-activated hydrochar .
MB800 . AB250 . NB250 . B250. The adsorption intensity or surface heterogeneity (1/n) for all hydrochars on both N and P adsorption was less than 1. This value below unity indicates that the adsorption process is primarily governed by the chemisorption process (Oginni et al. 2020). Although not fully understood, it is suggested that the surface area has little effect on adsorption compared to its elemental composition. Xue et al. (2009) reported that chemical composition of adsorbent led to simultaneous ligand exchange and chemical precipitation between adsorbent and PO 4 3À -P. Previous studies have also suggested that surface functionality may not greatly influence phosphate adsorption since biochars tend to be negatively charged (Zeng et al. 2013).

CONCLUSION
This study investigated anaerobically digested cattail hydrochars obtained via acetic acid or sodium hydroxide assisted HTC, various activation methods, and the N and P adsorption capacities of these hydrochars. When the HTC temperature increased from 200 to 300°C, the hydrochar yields decreased from 40.2% to 31.6%, 37.5% to 28.3%, and 35.7% to 22.7% for the use of water, acetic acid, and sodium hydroxide, respectively. The characterization showed that acetic acid or sodium hydroxide assisted HTC and activation with magnesium chloride or air could improve the surface properties. Acetic acid modification generated extensive oxygenated functional groups, while sodium hydroxide modification produced hydrochar with a high ratio of N/C and surface aromaticity. Treating hydrochar with magnesium chloride could impregnate nano-magnesium particles on the hydrochar, thereby improving the ability to remove N and P. Air activation of hydrochar resulted in a higher oxygen content. The NH 4 þ -N and PO 4 3À -P adsorption capacities of these hydrochars ranged from 92.6 to 122.4 mg/g and 1.6 to 15.8 mg/g, respectively. The adsorption capacity of hydrochars in swine wastewater is similar to the results of artificial wastewater. Despite the introduction of different surface functional groups, the hydrochars obtained by HTC assisted by acetic acid and sodium hydroxide did not significantly change the adsorption capacity of hydrochar. Hydrochar activated by using MgCl 2 and air substantially increased the adsorption capacity. The results suggested that Mg-nanoparticle dispersion and oxygen containing functional groups played a major role in adsorption than ion exchange and physisorption. Both NH 4 þ -N and PO 4 3À -P adsorption fit the Freundlich model well.