Phosphorus and Nitrogen Adsorption Capacities of Biochars Derived from Feedstocks at Di ﬀ erent Pyrolysis Temperatures

: This study investigates the P and NO 3 − adsorption capacities of di ﬀ erent biochars made from plant waste including rice straw (RSB), Phragmites communis (PCB), sawdust (SDB), and egg shell (ESB) exposed to a range of pyrolysis temperatures (300, 500 and 700 ◦ C). Results indicate that the e ﬀ ect of pyrolysis temperature on the physiochemical properties of biochar varied with feedstock material. Biochars derived from plant waste had limited adsorption or even released P and NO 3 − , but adsorption of P capacity could be improved by adjusting pyrolysis temperature. The maximum adsorption of P on RSB700, PCB300, and SDB300, produced at pyrolysis temperature of 700, 300 and 300 ◦ C, was 5.41, 7.75 and 3.86 mg g − 1 , respectively. ESB can absorb both P and NO 3 − , and its adsorption capacity increased with an increase in pyrolysis temperature. The maximum NO 3 − and P adsorption for ESB700 was 1.43 and 6.08 mg g − 1 , respectively. The less negative charge and higher surface area of ESB enabled higher NO 3 − and P adsorption capacity. The P adsorption process on RSB, PCB, SDB and ESB, and the NO 3 − adsorption process on ESB were endothermic reactions. However, the NO 3 − adsorption process on RSB, PCB and SDB was exothermic. The study demonstrates that the use of egg shell biochar may be an e ﬀ ective way to remove, through adsorption, P and NO 3 − from wastewater. on how biochars derived from plant waste and egg shell a ﬀ ect P and NO 3 − adsorption at varying pyrolysis temperatures. The objectives of this study therefore include: (1) to examine capacity of P and NO 3 − adsorption of di ﬀ erent biochars produced under di ﬀ erent pyrolysis temperatures; (2) to investigate the adsorption process of P and NO 3 − by the biochars. The results will o ﬀ er a theoretical and practical foundation for removal of P and NO 3 − from eutrophic or wastewaters using biochars.


Introduction
Discharges of phosphorus (P) and nitrate (NO 3 − ) into the natural environment from agricultural, industrial, and domestic wastewater have increased in many countries. The resulting nutrient enrichment and eutrophication of water-bodies has become a serious environmental concern around the world [1,2]. Under eutrophic conditions, rapid growth of organisms, especially algae, may be stimulated resulting in depletion of dissolved oxygen and deterioration of the aquatic environment [3]. In addition, human health could be affected by high levels of NO 3 − in water. For example, infant methemoglobinemia and several types of cancers could occur due to human uptake of excess

Materials
Four types of materials, i.e., rice straw, Phragmites communis, sawdust and egg shell were obtained from Nanjing University of Information Science and Technology, Nanjing, China. The materials were rinsed with water and air-dried, then ground and sieved to < 2.0 mm particles using a stainless grinding machine. The powdered biomass was tightly placed in a ceramic pot, and then pyrolyzed in a muffle furnace. The pyrolysis was programmed to drive temperature to 300 • C, 500 • C and 700 • C at a rate of 5 • C min −1 , respectively, and held at the peak temperature for 2 h before being cooled to room temperature. The biochars produced from rice straw, Phragmites communis, sawdust and egg shell under 300 • C, 500 • C and 700 • C were referred to as RSB300, RSB500, RSB700, PCB300, PCB500, PCB700, SDB300, SDB500, SDB700, ESB300, ESB500 and ESB700, respectively. All biochar samples were ground to pass through a 0.5 mm sieve prior to use.

Adsorption Kinetics Experiments
The stock solutions of P and NO 3 − -N were prepared using KH 2 PO 4 and KNO 3 , respectively.
Sorption kinetics were evaluated at room temperature (30 • C) and the initial pH for each sorption solution was adjusted to 7 prior to the experiments. The biochar (0.2 g) was added into 20 mL solutions containing 20 mg P or NO 3 − -N L −1 . Subsamples were taken after 30 min, 60 min, 180 min, 300 min, 420 min, 780 min, 1080 min and 1440 min, and shaken at 180 rpm in a mechanical shaker. The subsamples were then filtered with a syringe filter. The concentrations of NO 3 − -N and P in the filtrate were determined by ultraviolet spectrophotometry and ammonium molybdate spectrophotometry, respectively. The amount of P and NO 3 − adsorbed by the biochars was calculated by the following equation (Equation (1)).
where q t (mg g −1 ) is the amount of P and NO 3 − adsorbed by the biochars; C 0 and C t (mg L −1 ) are the initial and t time concentrations of the pollutants, respectively; V (L) is the volume of adsorption solution; and W (g) is the mass of biochar. The experimental results were fitted using three typical kinetic models (Pseduo-first-order Equation (2), Pseduo-second-order Equation (3) and Intraparticle diffusion Equation (4). ln(q e − q t ) = ln q e − k 1 t (2) where q e and q t (mg g −1 ) are the amounts of P and NO 3 − adsorbed by the biochars at the equilibrium time and at time t, respectively; k 1 (h −1 ), k 2 (g mg −1 h −1 ), and k 3 (mg g −1 h −0.5 ) are the rate constants of the corresponding model; and C (mg g −1 ) is a constant.

Adsorption Isotherm Experiment
Sorption isotherms of P or NO 3 − -N were determined using batch experiments in the centrifugal tube under the same conditions as above, and the concentration of P varied from 0 to 320 mg L −1 (0, 5, 10, 20, 40, 80, 160 and 320 mg L −1 ) or concentration of NO 3 − -N varied from 0 to 320 mg L −1 (0, 5, 10, 20, 40, 80, 160 and 320 mg L −1 ). After being shaken for 24 h, the final suspensions were centrifuged, filtered, and the supernatant solution was separated for analysis of P or NO 3 − . The concentration of P or NO 3 − was calculated Equation (5).
where q e is the equilibrium P or NO 3 − concentration in mg g −1 ; V is the volume of P or NO 3 − aqueous solution in L; W is the adsorbent mass in g; C 0 is the initial P or NO 3 − concentration in mg L −1 ; and C e is the aqueous P or NO 3 − concentration at equilibrium in mg L −1 .
Sorption isotherms were fitted to the Langmuir (Equation (6) and Freundlich (Equation (7)) equations to quantify the adsorption capacity of different biochars.
where q e (mg g −1 ) is the adsorption capacity; C e (mg L −1 ) is the equilibrium concentration after the adsorption or desorption; 1/n is the intensity of adsorption or affinity; K F (mg g −1 ) is the Freundlich adsorption constant; q max (mg g −1 ) is the maximum sorption capacity; K L (L mg −1 ) is a Langmuir constant.

Adsorption Thermodynamics
Sorption data of RSB700, PCB300, SDB300, and ESB700 using initial P or NO 3 − concentration (20,40,80,160 and 320 mg L −1 ) at a temperature range of 20, 30 and 40 • C were collected after 24 h equilibration time. Three parameters (Gibb's free energy change (∆G 0 ), enthalpy change (∆H 0 ) and entropy change (∆S 0 )) were calculated using the following equations. The thermodynamic equilibrium constant Kc was defined as (Equation (8)): where C 0 and C e (mg L −1 ) are the initial and equilibrium concentration of P or NO 3 − solution.
∆G 0 was calculated by the following equation (Equation (9)): where, T is temperature in K, R the ideal gas constant = 8.314 J mol −1 K −1 . ∆H 0 and ∆S 0 was calculated by the following equation (Equation (10)).

Analysis Method
The pH of biochars was measured by adding the biochars to deionized water at a mass/water ratio of 1:20 (PHS-3C). Each sample was analyzed in duplicate. The specific surface area and porosity properties of the biochars were measured by N 2 adsorption isotherms at 77 K with the Brunauer-Emmett-Teller (BET) method and by CO 2 isotherms at 273 K using a Quadrasorb Si-MP surface area analyzer. Zeta-potential measurements were performed at pH 7 with a potential analyzer (Zetasizer Nano ZS90, Malvern, UK).

Statistical Analysis
The average was calculated from three replicates of each experimental treatment using Origin Pro 8.0, and the results were indicated as mean ± standard deviation. The kinetics and sorption isotherms were fitted using Origin Pro 8.0, and R 2 values were used to compare the performance of different models. Statistical analysis was performed using SPSS 12.0. A one-way analysis of variance (ANOVA) was conducted for biochar characteristics. The tukey test was performed to detect the statistical significance of differences (p < 0.05) among means of treatments.

Characteristics of Biochar
According to Table 1, total volume, specific surface area and pH of RSB, PCB, SDB and ESB increased with pyrolysis temperatures, and the parameters were significantly higher at 700 • C than at 300 • C (p < 0.05). Data are means ± SD of n = 3. Different letters in the same column indicate significant differences in different pyrolysis temperature for each biochar (p < 0.05).
Zeta potential of PCB and ESB increased with pyrolysis temperatures ( Table 1). The zeta potential significantly decreased for RSB and SDB produced from pyrolysis at 500 • C than at 300 • C (p < 0.05). However, zeta potential significantly increased for RSB and SDB produced from pyrolysis at 700 • C than at 300 • C or 500 • C (p < 0.05). Results showed that the response of zeta potential of biochars on pyrolysis temperature varied with biochar type.
The SEM images of the studied biochars are shown in Figure 1, the more and uniform hollow channels for each biochar occurred with increasing pyrolysis temperature. Compared to RSB, PCB and SDB biochar, ESB biochar had less small pore size with relatively lower porosity ( Figure 1). isotherms were fitted using Origin Pro 8.0, and R 2 values were used to compare the performance of different models. Statistical analysis was performed using SPSS 12.0. A one-way analysis of variance (ANOVA) was conducted for biochar characteristics. The tukey test was performed to detect the statistical significance of differences (p < 0.05) among means of treatments.

Characteristics of Biochar
According to Table 1, total volume, specific surface area and pH of RSB, PCB, SDB and ESB increased with pyrolysis temperatures, and the parameters were significantly higher at 700 °C than at 300 °C (p < 0.05). Data are means ± SD of n = 3. Different letters in the same column indicate significant differences in different pyrolysis temperature for each biochar (p < 0.05).
Zeta potential of PCB and ESB increased with pyrolysis temperatures ( Table 1). The zeta potential significantly decreased for RSB and SDB produced from pyrolysis at 500 °C than at 300 °C (p < 0.05). However, zeta potential significantly increased for RSB and SDB produced from pyrolysis at 700 °C than at 300 °C or 500 °C (p < 0.05). Results showed that the response of zeta potential of biochars on pyrolysis temperature varied with biochar type.
The SEM images of the studied biochars are shown in Figure 1, the more and uniform hollow channels for each biochar occurred with increasing pyrolysis temperature. Compared to RSB, PCB and SDB biochar, ESB biochar had less small pore size with relatively lower porosity ( Figure 1).

Sorption Kinetics of Biochar
Adsorption or desorption of P on the biochars is shown in Figure 2. Release of P occurred in RSB300 and RSB500, PCB500 and PCB700, and SDB500 and SDB700 (Figure 2a-c), respectively. Similarly, PCB300 and SDB300 presented lower P adsorption capacity (Figure 2b,c): Novais et al. [23] showed that some biochars have very low or zero P adsorption, due to its behavior as "great anion", which prevents the adsorption of anions such as phosphates. Schneider and Haderlein [24] demonstrated that the dissolved organic matter released from the biochar pyrolysis at 200 °C when placed into the P solution competed for sorption sites and inhibited P sorption. However, our results show that P was adsorbed by ESB300, ESB500 and ESB700 (Figure 2d), P adsorption capacity is therefore dependent on biochar type and the rate of adsorption is affected by pyrolysis temperature.
Release of NO3 − occurred in RSB, PCB500, PCB700 and SDB700 (Figure 2e-g), respectively. However, adsorption of NO3 − by ESB increased with time, and increased with pyrolysis temperature ( Figure 2h). Therefore, ESB could adsorb both P and NO3 − . As shown in Figure 2d,h, when the initial P and NO3 − concentration was 20 mg/L, rapid adsorption on ESB was observed in the first 8h, which suggested that P and NO3 − in solution was impelled to adhere to the surface of biochar.

Sorption Kinetics of Biochar
Adsorption or desorption of P on the biochars is shown in Figure 2. Release of P occurred in RSB300 and RSB500, PCB500 and PCB700, and SDB500 and SDB700 (Figure 2a-c), respectively. Similarly, PCB300 and SDB300 presented lower P adsorption capacity (Figure 2b,c): Novais et al. [23] showed that some biochars have very low or zero P adsorption, due to its behavior as "great anion", which prevents the adsorption of anions such as phosphates. Schneider and Haderlein [24] demonstrated that the dissolved organic matter released from the biochar pyrolysis at 200 • C when placed into the P solution competed for sorption sites and inhibited P sorption. However, our results show that P was adsorbed by ESB300, ESB500 and ESB700 (Figure 2d), P adsorption capacity is therefore dependent on biochar type and the rate of adsorption is affected by pyrolysis temperature.   However, adsorption of NO 3 − by ESB increased with time, and increased with pyrolysis temperature ( Figure 2h). Therefore, ESB could adsorb both P and NO 3 − . As shown in Figure 2d,h, when the initial P and NO 3 − concentration was 20 mg/L, rapid adsorption on ESB was observed in the first 8h, which suggested that P and NO 3 − in solution was impelled to adhere to the surface of biochar.
To investigate adsorption mechanisms of P and NO 3 − , the sorption data were fitted with kinetic models, including the pseudo-first-order, pseudo-second-order and intra-particle diffusion model. According to Table 2, the pseudo-second-order kinetic model for adsorption P on RSB700, PCB300 and SDB300 showed the best fit to the experimental data with the highest R 2 in a range of 0.994-0.999. These results are similar to those of Elsa et al. [25], who demonstrated that the pseudo-second-order kinetic model fits the experimental data better than the pseudo-first-order kinetic model for P adsorption. However, the pseudo-first-order kinetic model for ESB showed the best fit to the experimental data with the highest R 2 in a range of 0.984-0.995 (Table 2). In this study, the intra-particle diffusion model didn't fit the data well with low R 2 in a range of 0.601-0.893 ( Table 2), indicating that intrapore diffusion does not dominate the adsorption process of P on the biochars. Table 2. Parameters of P adsorption kinetics of different biochars.

Sample Pseudo-First-Order
Pseudo-Second-Order Intra-particle Diffusion According to Table 3, the pseudo-second-order kinetic model for NO 3 − adsorption on PCB300, SDB300 and SDB500 showed the best fit to the experimental data with the highest R 2 in a range of 0.974-0.992. This suggests that NO 3 − adsorption process is mainly chemical adsorption. However, the pseudo-first-order kinetic model for NO 3 − adsorption on ESB showed the best fit to the experimental data with the highest R 2 in a range of 0.906-0.989. The results suggest that the intra-particle diffusion model didn't fit the data well with low R 2 in a range of 0.784-0.914 (Table 3). The pseudo-first-order model is widely used to describe reversible physical adsorption between the adsorbent and the adsorbate [26]. The pseudo-second-order model indicated that the adsorption of P onto calcium-flour biochar was a chemisorptions-dominated process [27]. Therefore, in this study, the pseudo-second-order model could be used to predict the kinetic process for P and NO 3 − sorption on PCB and SDB, which was a chemisorptions-dominated process. However, the pseudo-first-order model could be suitable to predict the kinetic process for P and NO 3 − sorption on EBS, which is a physisorption-dominated process.

Sorption Isotherms of Biochar
The amount of P adsorption by RSB700 increased with P concentration in the initial solution. Release of P from RSB300 and RSB500 occurred when P concentration was < 80 mg L −1 in the initial solution (Figure 3a). The amount of P adsorption by PCB300 and PCB500 increased with P concentration in the initial solution (Figure 3b). However, release of P from PCB700 occurred for different P concentrations in the initial solution (Figure 3b). Zhang et al. [19] showed that timber biochar and peanut shell biochar released P when the P concentration was < 100 mg P L −1 , but they retained 1-2% P when the P concentration in the solution was 200 mg P L −1 . It can be seen from Figure 3c that SDB300, SDB500 and SDB700 released P when the P concentration was < 5 mg L −1 , but they adsorbed P when P concentration was >80 mg L −1 . Therefore, the P adsorption capacity of the biochars was influenced by the P concentration in the initial solution. These results are similar to those of Chintala et al. [22], who demonstrated that P adsorption on biochars was significantly affected by initial P concentration and biochar types. According to Figure 3d, ESB300, ESB500 and ESB700 had a capacity to adsorb P, and ESB700 exhibited the highest P sorption capacity.
Water 2019, 11, x FOR PEER REVIEW 9 of 16 biochar and peanut shell biochar released P when the P concentration was < 100 mg P L −1 , but they retained 1-2% P when the P concentration in the solution was 200 mg P L −1 . It can be seen from Figure  3c that SDB300, SDB500 and SDB700 released P when the P concentration was < 5 mg L −1 , but they adsorbed P when P concentration was >80 mg L −1 . Therefore, the P adsorption capacity of the biochars was influenced by the P concentration in the initial solution. These results are similar to those of Chintala et al. [22], who demonstrated that P adsorption on biochars was significantly affected by initial P concentration and biochar types. According to Figure 3d, ESB300, ESB500 and ESB700 had a capacity to adsorb P, and ESB700 exhibited the highest P sorption capacity.  It can be seen from Figure 3e and 3f, RSB and PCB cannot adsorb NO3 − in the lower initial NO3 − concentration. The results are similar to those of Hale et al. [15], who found that cacao-shell derived biochar could not adsorb NO3 − . The amount of adsorption NO3 − on SDB and ESB increased with initial NO3 − concentration (Figure 3g,h). The order of NO3 − adsorption capacity on ESB was that of ESB700 > ESB500 > ESB300 (Figure 3h). However, the order of NO3 − adsorption on SDB was shown as following: SDB500 > SDB300 > SDB700 (Figure 3g). These results showed that the NO3 − adsorption capacity on biochars was also influenced by types of biochar and pyrolysis temperatures. RSB700, PCB300, PCB500, SDB300, SDB500, and ESB have the ability to adsorb P from P solutions, so their P adsorption isotherms were investigated to elucidate the adsorption mechanisms. The model parameters for P on biochars for both Langmuir and Freundlich models are presented in Table 4, which show that the Langmuir model fits the experimental data better (0.981 > R 2 > 0.852) than the Freundlich model (0.897 > R 2 > 0.728). Similarly, PCB300, SDB300, SDB500, ESB300, ESB500 and ESB700 have the ability to adsorb NO3 − . Therefore, their model parameters for NO3 − on biochars for both Langmuir and Freundlich models are presented in Table 5, which show that the Langmuir model fits the experimental data better (0.996 > R 2 > 0.919) than the Freundlich model (0.949 > R 2 > 0.850). Therefore, the Langmuir equation fitted the data better than the Freundlich equation for adsorption P and NO3 − (Tables 4 and 5). This is consistent with Tan et al. [28], who showed that the Langmuir equation had the best fit for the experiment. Elsa et al. [25] demonstrated that the Freundlich constant (KF) increased with the calcium content, which confirms the increase in adsorption intensity of the biochar due to the chemical reaction between phosphate ions and Ca 2+ . In this study, KF for P and NO3 − adsorption on ESB increased with pyrolysis temperature.   [15], who found that cacao-shell derived biochar could not adsorb NO 3 − . The amount of adsorption NO 3 − on SDB and ESB increased with initial NO 3 − concentration (Figure 3g,h). The order of NO 3 − adsorption capacity on ESB was that of ESB700 > ESB500 > ESB300 (Figure 3h). However, the order of NO 3 − adsorption on SDB was shown as following: SDB500 > SDB300 > SDB700 (Figure 3g). These results showed that the NO 3 − adsorption capacity on biochars was also influenced by types of biochar and pyrolysis temperatures. RSB700, PCB300, PCB500, SDB300, SDB500, and ESB have the ability to adsorb P from P solutions, so their P adsorption isotherms were investigated to elucidate the adsorption mechanisms. The model parameters for P on biochars for both Langmuir and Freundlich models are presented in Table 4, which show that the Langmuir model fits the experimental data better (0.981 > R 2 > 0.852) than the Freundlich model (0.897 > R 2 > 0.728). Similarly, PCB300, SDB300, SDB500, ESB300, ESB500 and ESB700 have the ability to adsorb NO 3 − . Therefore, their model parameters for NO 3 − on biochars for both Langmuir and Freundlich models are presented in Table 5, which show that the Langmuir model fits the experimental data better (0.996 > R 2 > 0.919) than the Freundlich model (0.949 > R 2 > 0.850). Therefore, the Langmuir equation fitted the data better than the Freundlich equation for adsorption P and NO 3 − (Tables 4 and 5). This is consistent with Tan et al. [28], who showed that the Langmuir equation had the best fit for the experiment. Elsa et al. [25] demonstrated that the Freundlich constant (K F ) increased with the calcium content, which confirms the increase in adsorption intensity of the biochar due to the chemical reaction between phosphate ions and Ca 2+ . In this study, K F for P and NO 3 − adsorption on ESB increased with pyrolysis temperature.  The maximum adsorption of P and NO 3 − (q max ) on ESB was lower in the biochar produced through pyrolysis at 300 • C than at 700 • C (Tables 4 and 5), indicating that the maximum adsorption of P and NO 3 − increased with pyrolysis temperature. Our results are similar to those of Zhang et al. [20], who found that biochar, produced from horse manure and bedding compost at pyrolysis of 200 • C, released the P and NO 3 − . However, the maximum adsorption of P (q max ) on PCB and SDB was lower in the biochar produced at pyrolysis temperature of 500 • C than at 300 • C. Therefore, the maximum adsorption of P on biochar was influenced by pyrolysis temperature. The maximum adsorption of P (q max ) on RSB700, PCB300, SDB300 and ESB700 was 5.407, 7.747, 3.859 and 6.084 mg g −1 , respectively. The maximum adsorption of P (q max ) on PCB300 was approximately two times that of SDB300. The maximum adsorption of NO 3 − (q max ) on PCB300, SDB300 and ESB700 was 0.443, 1.574 and 1.426 mg g −1 , respectively. Kameyama et al. [21] found that only 1.2 and 0.7 mg g −1 NO 3 − could be adsorbed by the bamboo powder charcoal and sugarcane bagasse derived biochar, respectively. In this study, the NO 3 − adsorption capacity was between 0.443 to 1.426 mg g −1 . Therefore, the capacity to absorb NO 3 − was lower but varied with the biochars from different feedstock materials.

Adsorption Thermodynamics of Biochars
The thermodynamics of P and NO 3 − adsorption on the biochars at 293, 303, and 313 K were analyzed (Figures 4 and 5). As shown in Table 6, the ∆H 0 value for RSB700, PCB300, SDB300 and ESB700 was 83.54, 44.63, 27.53 and 39.68 kJ mol −1 , respectively, indicating that P adsorption process was endothermic. The ∆S 0 value for RSB700, PCB300, SDB300 and ESB700 was 0.33, 0.19, 0.13 and 0.19 kJ mol −1 K −1 , respectively, indicating increased disorder and randomness of liquid-solid phase interaction at the biochar surface [29]. Zhang et al. [30] demonstrated that P tended to be adsorbed on the surface of biochar when the ∆S 0 values were positive. The ∆G 0 values were in a range of −10.48 to −18.41 kJ mol −1 indicating that the process of PO 4 3− adsorption onto the biochar was mainly spontaneous [31,32]. Furthermore, the ∆G 0 decreased with increasing adsorption reaction temperature, indicating a better P adsorption efficiency at a higher solution temperature [33].  The maximum adsorption of P and NO3 − (qmax) on ESB was lower in the biochar produced through pyrolysis at 300 °C than at 700 °C (Tables 4 and 5), indicating that the maximum adsorption of P and NO3 − increased with pyrolysis temperature. Our results are similar to those of Zhang et al. [20], who found that biochar, produced from horse manure and bedding compost at pyrolysis of 200 °C, released the P and NO3 − . However, the maximum adsorption of P (qmax) on PCB and SDB was lower in the biochar produced at pyrolysis temperature of 500 °C than at 300 °C. Therefore, the maximum adsorption of P on biochar was influenced by pyrolysis temperature. The maximum adsorption of P (qmax) on RSB700, PCB300, SDB300 and ESB700 was 5.407, 7.747, 3.859 and 6.084 mg g −1 , respectively. The maximum adsorption of P (qmax) on PCB300 was approximately two times that of SDB300. The maximum adsorption of NO3 − (qmax) on PCB300, SDB300 and ESB700 was 0.443, 1.574 and 1.426 mg g −1 , respectively. Kameyama et al. [21] found that only 1.2 and 0.7 mg g −1 NO3 − could be adsorbed by the bamboo powder charcoal and sugarcane bagasse derived biochar, respectively. In this study, the NO3 − adsorption capacity was between 0.443 to 1.426 mg g −1 . Therefore, the capacity to absorb NO3 − was lower but varied with the biochars from different feedstock materials.

Adsorption Thermodynamics of Biochars
The thermodynamics of P and NO3 − adsorption on the biochars at 293, 303, and 313 K were analyzed (Figures 4 and 5). As shown in Table 6, the ΔH 0 value for RSB700, PCB300, SDB300 and ESB700 was 83.54, 44.63, 27.53 and 39.68 kJ mol −1 , respectively, indicating that P adsorption process was endothermic. The ΔS 0 value for RSB700, PCB300, SDB300 and ESB700 was 0.33, 0.19, 0.13 and 0.19 kJ mol −1 K −1 , respectively, indicating increased disorder and randomness of liquid-solid phase interaction at the biochar surface [29]. Zhang et al. [30] demonstrated that P tended to be adsorbed on the surface of biochar when the ΔS 0 values were positive. The ΔG 0 values were in a range of −10.48 to −18.41 kJ mol −1 indicating that the process of PO4 3− adsorption onto the biochar was mainly spontaneous [31,32]. Furthermore, the ΔG 0 decreased with increasing adsorption reaction temperature, indicating a better P adsorption efficiency at a higher solution temperature [33].       (c) (d) Figure 5. Amount of NO3 − on RSB500 (a), PCB300 (b), SDB500 (c) and ESB300 (d) under different temperature, respectively. Data are means ± SD of n = 3. Table 6. Thermodynamic parameters for P and NO3 − adsorption.     [33]. Results show that NO 3 − adsorption process on the biochars produced from feedstocks rich in cellulose, hemicellulose and lignin was an exothermic reaction, but an endothermic reaction occurred for the biochar produced from egg shell.

Effect of Characteristic of Biochar on Capacity of Nitrate and Phosphate Adsorption
Cellulose, hemicellulose, and lignin are essential constituents of plant cell walls. In this study, RSB, PCB and SDB derived from the plants waste showed limited adsorption of, or even released, NO 3 − and P. These findings agree with Zhang et al. [34] who found that the pristine biochar surface is negatively charged, and thus cannot easily adsorb the negatively charged NO 3 − and P. Plant-derived biochar used in this study had a low zeta potential (−16.6 to −41.9 mV), meaning that electrostatic repulsion between the negatively charged surface sites and electronegative phosphate species resulted in the lower phosphate adsorption [35]. Yang et al. [36] demonstrated that positive zeta potential is desirable to the adsorption of anion by electrostatic attraction and in this study egg shell biochar (ESB) had a positive zeta potential in the range −2.9 to −4.9 mV. The NO 3 − and P adsorption capacity of egg shell biochar was influenced by surface area and porosity characteristic. The SEM images of ESB showed that there was more porosity of ESB produced at 700 • C than at 300 • C (Figure 1). The P and NO 3 − adsorption capacities were higher in ESB700 than in ESB300, which was related to more porous structure and higher surface area in ESB700 than in ESB300 (Table 1). These findings agreed with those of Yin et al. [37], who demonstrated that the porous structure and higher surface area could contribute to better PO 4 3− and NO 3 − adsorption. Therefore, the higher surface area of egg shell biochar seems to improve adsorption of NO 3 − and P.
The NO 3 − and P adsorption capacity of biochars could be improved by adjusting pyrolysis temperature. Many methods have been used to improve NO 3 − and P adsorption capacity. It was reported that the dashed activated carbon could adsorb 9.84 mg NO 3 − g −1 [38]. The NO 3 − adsorption on the La-modified biochar was 8.81 mg g −1 , which was considerably higher than that on the pristine biochar (2.81 mg g −1 ) [39]. The Mg-modified sugar beet tailing biochar showed a P adsorption capacity of 6.67 mg g −1 [35]. The sesame straw biochar activated by ZnCl 2 showed the highest phosphorus adsorption capacity of 9.39 mg g −1 [40]. The poplar chips biochar modified by Al showed its high P adsorption capacity of 43.98 mg g −1 [37]. Results of this study showed that the NO 3 − and P adsorption capacity could be enhanced by adjusting pyrolysis temperature. For example, the maximum sorption of P on RSB, produced at pyrolysis temperature of 700 • C, was 5.40 mg g −1 . Compared with RSB300 and RSB500, RSB700 with higher P adsorption capacity could be attributed to higher surface area and total volume (Table 1). Similarly, the maximum sorption of NO 3 − on SDB, derived from pyrolysis at 300 • C, was 1.574 mg g −1 . Iida et al. [38] demonstrated that the NO 3 − adsorption process would be restricted under basic conditions because the OH − competes with NO 3 − for the adsorption sites on