Adsorption Characteristics of Phosphate Ions by Pristine, CaCl 2 and FeCl 3 -Activated Biochars Originated from Tangerine Peels

: This study has evaluated the removal efficiencies of phosphate ions (PO 43 − ) using pristine (TB) and chemical-activated tangerine peel biochars. The adsorption kinetics and isotherm presented that the enhanced physicochemical properties of TB surface through the chemical activation with CaCl 2 (CTB) and FeCl 3 (FTB) were helpful in the adsorption capacities of PO 43 − (equilibrium adsorption capacity: FTB (1.655 mg g − 1 ) > CTB (0.354 mg g − 1 ) > TB (0.104 mg g − 1 )). The adsorption kinetics results revealed that PO 43 − removal by TB, CTB, and FTB was well fitted with the pseudo-second-order model (R 2 = 0.999) than the pseudo-first-order model (R 2 ≥ 0.929). The adsorption isotherm models showed that the Freundlich equation was suitable for PO 43 − removal by TB (R 2 = 0.975) and CTB (R 2 = 0.955). In contrast, the Langmuir equation was proper for PO 43 − removal by FTB (R 2 = 0.987). The PO 43 − removal efficiency of CTB and FTB decreased with the ionic strength increased due to the compression of the electrical double layer on the CTB and FTB surfaces. Besides, the PO 43 − adsorptions by TB, CTB, and FTB were spontaneous endothermic reactions. These findings demonstrated FTB was the most promising method for removing PO 43 − in waters.


Introduction
With a significant increase in the amount of nutrients introduced to the water system due to the rapid industrialization and recent population growth, increasing water pollution hinders effective water quality management [1,2]. Nutrients are divided into point and non-point sources, depending on the primary source of inflow. Although most effluents from domestic sewage treatment plants and livestock wastewater treatment plants meet the water quality criteria, agricultural drainage water, a representative non-point source, significantly affects the discharge of nutrients because the increased amounts of compost or fertilizer used to improve agricultural production [3,4]. This oversupply of nutrients into the water system through point and non-point sources increases eutrophication causes algal blooms. It reduces the amount of dissolved oxygen in the aquatic ecosystem, thereby causing the deaths of aquatic organisms [5]. Among the main components of nutrients, nitrate ions (NO3 − ) are required in relatively large quantities for algal growth, but phosphate ions (PO4 3− ) are a limiting factor that can promote algal growth even when present in small amounts. PO4 3− can lead to blue-green algal blooms, leading to renal failure through toxicity [6,7].
The anaerobic anoxic aerobic (A2O) process is typically applied for the biological treatment of PO4 3− . Although this process requires no chemical injection and generates little sludge compared to the amount of phosphate removed [8], the results are significantly affected by the operating conditions. Besides, A2O is not suitable for strict water quality criteria because it cannot protect microorganisms against toxic chemicals present in the influent water. Moreover, microorganisms in the A2O process are subject to more significant technical limitations in treatment efficiency compared with the physicochemical treatment process [6]. There are commonly used physicochemical treatment processes (e.g., Electrodialysis, membrane filtration, coagulation, precipitation, and adsorption) to remove PO4 3− from water and wastewater treatment plants [9]. These techniques are not appropriate to the full-scale water and wastewater treatment plants due to the high operating costs and energy consumption [10], and the generated sludge may cause secondary environmental pollution in subsequent treatment processes [11]. However, the adsorption process has been used in the water and wastewater treatment plants due to low operation and maintenance costs [12]. The various types of PO4 3− adsorbents, such as clay minerals [13], fly ashes [14], and metal oxides [15], have been investigated. Das et al. have demonstrated the high PO4 3− adsorption on clay minerals, including layered double hydroxides [13]. Chen et al. reported the efficient PO4 3− adsorption on fly ashes [14]. Zhang et al. have fabricated activated carbon fiber with metal oxides for PO4 3− adsorption [15]. Despite the advantage of these adsorbents, including high efficiency and environment-friendly properties, they were difficult to be applied the full-scale wastewater treatment plant for PO4 3− adsorption due to the high treatment cost (e.g., recycle and regeneration). Therefore, it was necessary to develop an alternative adsorbent to remove PO4 3− in water [16].
Biochars, alternative adsorbents, are carbon-rich substances obtained through biomass pyrolysis, such as fruit peels and rice bran, under oxygen-limited conditions. The utilization of biochar shows significant environmental advantages in reducing greenhouse gas emissions and resource recycling technology of agriculture and food residues [17]. The agricultural residues represent an important potential source of reusable products [18,19]. Tangerine peels are common agricultural residues in Korea. The amount of tangerines produced in Jeju Island was 125,343 tons as of 2005, representing approximately 18% of the total global tangerine production. Annually, over 55,000 tons have been discarded as tangerine peels [20]. Tangerine peels can be highly applicable as a raw material for biochars because they are mainly composed of pectin, hemicellulose, and cellulose substances [21,22]. The negative surface charge of pristine biochar has limited their adsorption affinity towards anions, including PO4 3− [23,24]. Therefore, their adsorption capacities might be considerably enhanced after modification, including chemical activation, surface functionality modification, and biochar impregnated with metals (e.g., CaCl2, MgCl2, FeCl3, and AlOOH) [24][25][26][27][28]. Fang et al. have demonstrated the high PO4 3− adsorption on MgCl2 modified ground corn biochar [27]. Zhang and Gao. have reported the efficient PO4 3− adsorption on AlOOH modified cottonwood biochar [24]. Despite the effective adsorption capacities of metal-loaded biochar for PO4 3− , these biochars preparation demanded high energy for pyrolysis, and the adsorption ability of biochars reduced due to coalescence with water [24,27]. CaCl2 and FeCl3, which are a type of chemical coagulants, are commonly used in the adsorption of PO4 3− in water. Thus, the modification of biochars with CaCl2 and FeCl3 could significantly improve the adsorption capacities of PO4 3− in water.
The primary purpose of this study is to evaluate the effect of pretreatment with CaCl2 and FeCl3 on the PO4 3− removal of biochars made from tangerine peels. Thus, the effects of various conditions, such as the biochar dosage, pH, ionic strength, and temperature, on PO4 3− removal were evaluated using the pristine tangerine peel biochar (TB), and CaCl2 (CTB) and FeCl3 (FTB) activated tangerine peel biochars. In addition, the PO4 3− adsorption mechanisms of TB, CTB, and FTB were investigated through adsorption kinetics and adsorption isotherm models.

Preparation of Tangerine Peel Biochars
Tangerine peels were purchased from a local food store on Jeju Island (Jeju-do, Korea). After dried tangerine peels were crushed to 0.5-1.0 mm using a blender, they were several rinsed (ten times) with DI water to remove impurities and then dried in an oven at 105 °C for 12 h. The crushed tangerine peels were immersed in 200 mL solutions of 1 M CaCl2 and 1 M FeCl3, respectively. They were then stirred at 80 °C for 1 h and dried in an oven at 105 °C for 24 h. The pristine and chemical activated tangerine peels were pyrolyzed at 800 °C for 1 h using a tubular furnace (PyroTech, Namyangju, Gyeonggi-do, Korea) under N2 gas (the flow rate = 0.25 L min −1 ) atmospheric conditions (heating rate = 5 °C min −1 ) [29]. After cooling to room temperature (20 ± 0.5 °C), the fabricated tangerine peel biochars were rinsed using DI water until no impurities were observed and dried in an oven at 80 °C for 24 h. The dried tangerine peel biochars (i.e., TB, CTB, and FTB) were sieved to obtain a homogenized particle size of 150 µm and then stored in a desiccator prior to use.

Characteristics of Tangerine Peel Biochars
Total carbon (C), nitrogen (N), and hydrogen (H) contents of the pristine and chemical activated tangerine peel biochars were analyzed using a CHN element analyzer (Flash 2000, Thermo Fisher, Waltham, MA, USA). The average pore size (nm) and specific surface area (m 2 g −1 ) were measured using a Brunauer-Emmett-Teller (BET; BELSORP-mini II, Microtrac BEL, Osaka, Japan) analyzer. An X-ray diffractometer (XRD; D/Max-2500, Rigaku, Tokyo, Japan) was used to analyze the surface crystallinity of TB, CTB, and FTB. The surface morphologies of TB, CTB, and FTB were observed using a ultra-high resolution scanning electron microscope (UH-SEM; S-4800, Hitachi, Tokyo, Japan), and the atomic-resolution chemical mapping of calcium and iron ions were identified using energy-dispersive X-ray spectroscopy (EDX; Link ISIS 300, Oxford Instruments, Abingdon, UK).

Optimal Dosage
The adsorption of PO4 3− was examined to determine the optimal adsorbent dosages of TB, CTB, and FTB. Each adsorbent dosage (TB = 0.2-2.0 g L −1 ; CTB = 0.2-12 g L −1 ; FTB = 0.2-2.0 g L −1 ) was added to Erlenmeyer flasks containing 25 mL of the PO4 3− solution (initial concentration = 1 mg L −1 , pH = 7.0) The sample solutions were stirred at 25 °C and 150 rpm for 24 h using a shaking incubator (VS-8480, Vision Scientific, Daejeon-Si, Korea). Upon completing the adsorption experiment, the sample solutions were filtered using a glass fiber filter (GF/F, Whatman, Maidstone, UK) with a nominal pore size of 0.7 µm to remove adsorbents. The PO4 3− concentration was analyzed at UV absorbances of 880 nm using the ascorbic acid method (UV-Vis Spectrophotometer, UV-1280, Shimadzu, Kyoto, Japan) [30]. The experiment was performed in triplicate to minimize errors.

Adsorption Kinetics
The adsorption kinetics was conducted by adding the optimal dosage of each TB, CTB, and FTB to Erlenmeyer flasks containing 25 mL of the PO4 3− solution (initial concentration = 1 mg L −1 , pH = 7.0). The sample solutions were stirred at 150 rpm during a certain period (0.5-48 h) at 25 °C in a shaking incubator. After the adsorption kinetics experiment, the sample solutions were filtered using GF/F. The concentrations of PO4 3− at the initial and equilibrium states were measured using a UV-Vis spectrophotometer. The experiment was performed in triplicate to minimize errors. All adsorption experiments are repeated three times to minimize errors. The amount of PO4 3− adsorbed per unit mass of the TB, CTB, and FTB at equilibrium, Qe (mg g −1 ), was calculated using the following equation (1): where V is the volume of the solution (L). C0 and Ce are the initial and equilibrium concentrations of PO4 3− solution (mg L −1 ), and M (g) is the mass of the used adsorbent. The PO4 3− removal efficiency was calculated using equation (2): The PO4 3− adsorption characteristics and adsorption capacity of each TB, CTB, and FTB were investigated using the following equations (3) and (4) [31]: Pseudo-first-order model: Pseudo-second-order model: where Qt (mg g −1 ) is the amount of the adsorbed PO4 3− on the TB, CTB, and FTB at the time t, t (min) is the adsorption time. k1 (min −1 ) is the constant of the pseudo-first-order model and k2 (g mg −1 •min) is the constant of the pseudo-second-order model.

Adsorption Isotherm
To investigate the adsorption isotherm of PO4 3− by TB, CTB, and FTB, the adsorption isotherm experiment was performed by adjusting the PO4 3− concentrations (0.5-10 mg L −1 ) and adding each optimal adsorbent dosage under controlled conditions (agitation speed = 150 rpm, contact time = 24 h, pH = 7.0, and temperature = 25 °C). The adsorption isotherm results were analyzed using the Langmuir isotherm and Freundlich isotherm models [32].
where Qmax (mg g −1 ) is the maximum adsorption capacity in the Langmuir isotherm model, and KL (L mg −1 ) is the equilibrium constant of the linearized Langmuir isotherm model. RL = 1/(1 + KLC0), derived from KL, can be used to compare the adsorption affinity of Langmuir isotherms [33]: where KF (mg 1-1/n L 1/n g −1 ) is the Freundlich isotherm adsorption constant related to the relative maximum adsorption capacity, and n is the dimensionless adsorption intensity.

Effects of pH and Ionic Strength
The effects of pH and ionic strength on the adsorptions of the PO4 3− by the TB, CTB, and FTB were evaluated by adjusting solution pH (pH = 3.0-9.0) and ionic strengths (ionic strength = 0-0.5 M) using 0.1 N HCl and 0.1 N NaOH, and NaCl, respectively (initial concentration of PO4 3− solution = 1 mg L −1 , agitation speed = 150 rpm, contact time = 24 h). The removal efficiencies of PO4 3− using TB, CTB, and FTB were calculated by equation (2).

Effects of Temperature
The effects of the temperature of the solution on the PO4 3− removal efficiency of the TB, CTB, and FTB were performed under various temperature (15-35 °C) conditions (initial concentration of PO4 3− solution = 1 mg L −1 , agitation speed = 150 rpm, contact time = 24 h, and pH 7.0). The removal efficiencies of PO4 3− using TB, CTB, and FTB were followed by Equation (2).
The thermodynamic parameters of the PO4 3− adsorption are calculated using the following equations (7)-(9) [34]: where Kd (L g −1 ) is the partition coefficient. ∆G° in (kJ mol −1 ), ∆H° in (kJ mol −1 ), and ∆S° in (J mol −1 •K) are the Gibbs free energy, enthalpy, and entropy, respectively. R is the ideal gas constant (8.314 J mol −1 •K), and T is the absolute temperature (K). ∆H° and ∆S° were calculated as the slope and intercept in the linear graph of ln Kd and 1/T, respectively.

Elemental Composition and Functionality Analyses
The elemental composition (i.e., C, H, O, and N) and surface properties (i.e., specific surface area pore volume and average pore size) of TB, CTB, and FTB associated with the adsorption capacity of PO4 3− are presented in Table 1 These results showed that FTB contained relatively less aromatic functional groups than those of TB and CTB [36]. Furthermore CTB and FTB exhibited larger specific surface areas (TB = 9.21 m 2 g −1 ; CTB = 342.11 m 2 g −1 ; 558.71 m 2 g −1 ), larger pore volumes (TB = 0.01 cm 3 g −1 ; CTB = 0.36 cm 3 g −1 ; FTB = 0.18 cm 3 g −1 ), and smaller average pore sizes (TB = 6.07 nm; CTB = 3.67 nm; FTB = 3.64 nm) compared with TB, indicating that activation process with CaCl2 and FeCl3 was effective in improving the physicochemical characteristics of the tangerine peel biochars related to PO4 3− adsorption [37]. The functional groups of TB, CTB, and FTB are revealed by FT-IR analysis (Figure 1). The main differences between TB and chemical activated TB (i.e., CTB and FTB) are the existence of C=O stretching of esters and -COO carboxylates. These functional groups might enhance the adsorption capacities of the phosphate ions using CTB and FTB by working as an electron acceptor [38,39].  Figure 3 shows the EDX mapping images of the TB, CTB, and FTB surfaces. The surface of TB was mostly composed of carbon (Figure 3a), whereas calcium and iron salts were evenly distributed on the surfaces of CTB and FTB (Figures 3b,c). Moreover, the results of EDX mapping were in good agreement with the atomic percentage of elements in TB, CTB, and FTB (Table 2). These observations indicate that calcium and iron salts were successfully impregnated in the surface of the tangerine peel biochars through pretreatment with CaCl2 and FeCl3.   The crystallinities of TB, CTB, and FTB were analyzed using XRD (Figure 4). The XRD peaks of TB related to graphite and quartz (SiO2) were found at 2θ = 23° and 43°, respectively [40]. The XRD peaks of CTB and FTB related to calcium and iron species (e.g., CaCO3, MgFe2O4) were found (CaCO3 at 2θ = 35°, 57°, and 65°; MgFe2O4 at 2θ = 30°, 35°, 43°, 57°, and 63°) [41]. These findings were in good agreement with the SEM-EDX analysis results of TB, CTB, and FTB.

Effects of Tangerine Peel Biochar Dosage
The adsorbent dosage is one of the critical factors which affect the adsorption of PO4 3− . Figure 5 presents the effects of the dosages of TB, CTB, and FTB on the removal efficiency of PO4 3− . In the case of TB, the removal efficiency of PO4 3− decreased as the adsorbent dosage was increased beyond 0.6 g•L −1 . These results indicated that the decreased adsorption capacity of TB for PO4 3− was caused by reducing the total number of binding sites on TB surfaces due to the aggregation of TB particles as increasing adsorbent dosage [42]. However, the removal efficiencies of the PO4 3− by CTB and FTB increased with the dosage increase. These results indicated that the activated binding sites of the adsorbents capable of PO4 3− adsorption increased with increasing dosage [43]. Furthermore, FTB was more effective in removing PO4 3− than that of CTB because the binding capacity of iron salts is higher than that of calcium [44]. Based on these experiments on the PO4 3− removal efficiency according to the TB, CTB, and FTB dosages, 0.6 g•L −1 was selected as the optimal dosage and applied to subsequent experiments.  Figure 6 shows the adsorption kinetics of PO4 3− by TB, CTB, and FTB. The adsorption process of PO4 3− is comprised of fast and slow reaction stage. The fast adsorption reaction was completed in about 0.5 h for TB, CTB, and FTB as the activated sites on the surface of the biochars were saturated. The fast adsorption reaction was completed in about 0.5 h for TB, CTB, and FTB as the activated sites on the surface of the biochars were saturated. The TB (Qe,exp = 0.104 mg g −1 ) and CTB (Qe,exp = 0.354 mg g −1 ) with relatively low removal efficiency compared to FTB, adsorption equilibrium was reached after 2 h. However, the adsorption equilibrium of FTB (Qe,exp = 1.655 mg g −1 ) was completed in 18 h due to many activated sites on the surface [45]. Table 3 presents the results of calculating the constant and correlation coefficient of adsorption kinetics. The adsorption of PO4 3− by TB, CTB, and FTB was well fitted for the pseudo-second-order model (R 2 = 0.999) than the pseudo-first-order model (R 2 ≥ 0.929). These results indicated that the adsorption of TB, CTB, and FTB is caused by chemical adsorption [46]. Table 3. The kinetic parameters for the removal of the PO4 3− using TB, CTB, and FTB.

Adsorption Isotherms
The adsorption behaviors of PO4 3− by the TB, CTB, and FTB were examined using the Langmuir and Freundlich adsorption isotherm models ( Table 4). The adsorption of PO4 3− by TB was well fitted to the Freundlich isotherm model with the high R 2 values (R 2 of Langmuir isotherm = 0.887; R 2 of Freundlich isotherm = 0.975). This is evidence that the multilayer adsorption played a critical role in removing the PO4 3− toward the heterogeneous surfaces of the TB [47]. For the CTB and FTB, the adsorption of PO4 3− followed both Langmuir (R 2 of CTB = 0.889; R 2 of FTB = 0.987) and Freundlich (R 2 of CTB = 0.955; R 2 of FTB = 0.912) isotherm models. These observations could explain that the chemical activation with CaCl2 and FeCl3 might change the adsorption mechanism (i.e., multilayer adsorption → monolayer adsorption) of PO4 3− by the TB. A similar result was previously observed for the removal of the pharmaceuticals with NaOH-activated biochars [48]. The adsorption affinities of PO4 3− to the TB, CTB, and FTB were evaluated using the n values (dimensionless adsorption intensity) of the Freundlich isotherm model: (i) n > 1.0 (favorable), (ii) n = 1.0 (linear), and (iii) n < 1.0 (unfavorable) [49]. The adsorption of PO4 3− by TB (n value = 0.766) was unfavorable, whereas the adsorptions of PO4 3− by CTB (n value = 1.523) and FTB (n value = 7.530) were favorable. The RL value (maximum adsorption capacity; RL = 1/(1 + KLC0)) of the Langmuir isotherm model: (i) RL = 0 (irreversible), (ii) 1 > RL > 0 (favorable), (iii) RL = 1 (linear), and (iv) RL > 1 (unfavorable), was assessed to the adsorption affinities of PO4 3− toward TB, CTB, and FTB [50]. The adsorption of PO4 3− by FTB (RL = 0.209) followed the Langmuir isotherm model and seemed to be favorable for the monolayer adsorption [51]. Moreover, these results are comparable to the maximum adsorption capacity (mg g −1 ) calculated using different adsorbents as shown in Table 5.  Figure 7 illustrates the effect of pH (pH = 3-9) on the adsorption of PO4 3− using TB, CTB, and FTB. It was presented that the removal efficiency of PO4 3− by TB, CTB, and FTB was not significantly affected by the pH change (removal efficiency of TB = 10.9-12.1%; removal efficiency of CTB = 25.1-29.8%; removal efficiency of FTB = 93.3-99.4%). These results indicated that TB, CTB, and FTB could be used to effectively remove PO4 3− from wastewater with a wide range of pH [54].

Effects of Ionic Strength on Adsorption of PO4 3−
The effects of ionic strength (ionic strength = 0-0.5 M) on the adsorption of PO4 3− by TB, CTB, and FTB are shown in Figure 8. The removal efficiency of PO4 3− by TB was not significantly affected by the ionic strength change (the removal efficiency of PO4 3− = 9.7%→10.5%). However, the removal efficiencies of PO4 3− using the CTB and FTB were gradually decreased with increasing ionic strengths (CTB: the removal efficiency of PO4 3− = 25.2%→14.8%; FTB: the removal efficiency of PO4 3− = 98.8%→55.1%). These observations suggested that increases in ionic strength might reinforce the electrostatic repulsion between PO4 3− and adsorbent surfaces, and activated adsorption sites on the surface might be reduced due to the compression of the electrical double layer on the adsorbent surfaces [55].

Effects of Temperature and Thermodynamic Analysis
The effects of the temperature on the removal efficiencies of the PO4 3− by TB, CTB, and FTB are compared in Figure 9 (temperature = 15-35 °C). The adsorption of PO4 3− on the CTB gradually increased with increasing temperature ( Figure 9, the removal efficiency of CTB = 18.2%→37.5%). A possible explanation for these results is that increasing temperature cause more strong intermolecular motion and PO4 3− diffusion rate to the surface of CTB, which promoted the adsorption of PO4 3− on the CTB [56]. However, the removal efficiencies of PO4 3− by TB and FTB were not significantly affected by the temperature change (removal efficiency of TB = 7.3%-7.9%; removal efficiency of FTB = 98.2%-100.0%). Table 6 shows the values of the thermodynamic parameters (∆G°, ∆H°, and ∆S°) for PO4 3− removal by TB, CTB, and FTB according to the temperature (15-35 °C). The ∆G° < 0 and ∆H° > 0 suggested that the adsorption of PO4 3− on the TB, CTB, and FTB was a spontaneous and endothermic reaction [57,58]. Furthermore, ∆S° > 0 indicated that the adsorption of PO4 3− on the TB, CTB, and FTB was irreversible, which was conducive to the adsorption stability [56].

Conclusions
This study verified that pretreatment with CaCl2 and FeCl3 could improve the surface characteristics of tangerine peel biochars related to the adsorption behaviors of PO4 3− . The FTB might more effectively remove the PO4 3− (Qe, exp = 1.655 mg g −1 ) than TB (Qe, exp = 0.104 mg g −1 ) and CTB (Qe, exp = 0.354 mg g −1 ) due to the considerable enhancement of the Temperature (℃ ) 15 25 35 Removal efficiency (%) physicochemical characteristics (specific surface area and surface characteristics). The removal efficiencies of PO4 3− by TB (R 2 = 0.975) and CTB (R 2 = 0.955) were more suitable for the Freundlich adsorption model (multilayer adsorption) and the FTB was well fitted to the Langmuir adsorption model (R 2 = 0.987, monolayer adsorption). Furthermore, the thermodynamic analysis presented that the adsorption of PO4 3− for the FTB was more spontaneously endothermic than that for the TB and CTB under various pH and ionic strength conditions. These results are evidence that the chemical activation with FeCl3 might be a promising option to make the pristine tangerine peel biochar practically more relevant for the removal of PO4 3− in the aqueous solutions.